Membrane‐tethered Notch1 exhibits oncogenic property via activation of EGFR–PI3K–AKT pathway in oral squamous cell carcinoma

Yang Zheng1,2* | Zhao Wang1,3* | Xianbin Xiong1,2* | Yi Zhong1,4 | Wei Zhang1,4 | Yibo Dong1,2 | Jialiang Li1,2 | Zaiou Zhu1,2 | Wei Zhang1 | Heming Wu1,2 | Wenyi Gu3 | Yunong Wu1,2 | Xiang Wang5 | Xiaomeng Song1,2


Notch proteins are highly conserved cell surface receptors which play essential roles in cellular differentiation, proliferation, and apoptotic events at all stages of development. Recently, NOTCH1 mutations have been extensively observed in oral squamous cell carcinoma (OSCC) and are hinted to be Notch1‐inactivating mutations. However, little is known about the biological effect of these reported mutations in OSCC. To mimic the inactivation of Notch1 due to inappropriate mutations and to determine the potential mechanisms, we utilized wild‐type Notch1 vectors (Notch1WT) or mutant Notch1 vectors (Notch1V1754L) to transfect into OSCC cell lines. Membrane‐tethered Notch1 induced by mutation was analyzed by immunofluorescence staining. γ‐Secretase inhibitor PF‐03084014 was utilized to determine the phenotype in the absence of endogenous Notch1 activation. Here we demonstrated that membrane‐tethered Notch1 inactivated the canonical Notch1 signaling and oncogenic phenotypes were identified by promoting cell proliferation and invasion and by inducing epithelial‐to‐mesenchymal transition in cells. The γ‐secretase inhibitor PF‐03084014 also showed distinct oncogenic property after treatment. Importantly, both membrane‐tethered Notch1 and PF‐03084014 inhibitor activated the epidermal growth factor receptor (EGFR)–phosphoinositide 3‐kinase (PI3K)– protein kinase B (AKT) signaling pathway, which has been confirmed as an overwhelming modulator in OSCC. This was the first time that we clearly simulated the mutated Notch1 activities and determined the oncogenic phenotypes of membrane‐tethered Notch1. Compared with wild‐type Notch1, membrane‐tethered Notch1 was strongly associated with activated EGFR–PI3K–AKT signaling pathway.

AKT, EGFR, EMT, membrane‐tethered Notch1, OSCC


Oral squamous cell carcinoma (OSCC), one of the most common cancers worldwide, is notorious for poor prognosis, which reflects the inclination of OSCC to present as clinically advanced disease upon diagnosis (Izumchenko et al., 2015; Song et al., 2014; Zhao et al., 2016; Zheng et al., 2017). With the emergence of next‐generation sequencing technology, functional driver mutations in mutated genes have been validating and identifying lately (Liu et al., 2016; Mardis, 2008; Stransky et al., 2011; Zanaruddin et al., 2013). A better conception of the molecular mechanisms underlying the develop- ment of OSCC driven by aberrant mutations may add to the treatment and prognosis of this disease.
Notch proteins are highly conserved cell surface receptors which play essential roles in cellular differentiation, proliferation, and apoptotic events at all stages of development (Jeffries & Capobianco, 2000; Sun et al., 2014; Zheng et al., 2017). Notch1 receptor is synthesized as a single 300‐kDa polypeptide in the endoplasmic reticulum (Figure 1a). Following S1 cleavage, two noncovalently associated extracellular and transmembrane subunits mediated by heterodimerization (HD) domain are presented on the cell surface (Malecki et al., 2006). Ligands of the Delta and Jagged families presented on adjacent cells can interplay with the extracellular domain of Notch1 (NECD). After NECD is cleaved by a disintegrin andmetallo- proteinase (ADAM)/tumournecrosis factor α‐converting enzyme (TACE) metalloproteases, the Notch1 intracellular domain (NICD) is released from the membrane by γ‐secretase proteolysis (S3 cleavage; De Strooper et al., 1999). NICD is consecutively released from the plasma membrane and translocates into the nucleus, where it acts as an activator with transcription factors of the C promoter binding factor‐1– suppressor of hairless/lag‐1 (CSL) family to modulate transcription of downstream target genes including HES and HEY families (Le Borgne, 2006; Malecki et al., 2006; McGill & McGlade, 2003; Pei & Baker, 2008; Sharma, Rangarajan, & Dighe, 2013; Sun et al., 2014; Yoshida et al., 2013). Although S1–S3 cleavages are generally considered as the canonical proteases involved in Notch1 activation, increasing data have shown that some molecules could process Notch1 in a noncanonical way (Ma et al., 2014; Traustadóttir et al., 2016).
There is considerable debate concerning what the Notch1 functions and what proteins serve as effectors of the Notch1 signal (Jeffries & Capobianco, 2000). As several Notch genes have been shown to be proto‐oncogenes in mammalian cells, activation of Notch1 has been associated with tumorigenesis in several human cancers (Palomero et al., 2007). Nevertheless, systematic functional studies that elucidate the exact role of the Notch1 in OSCC are hard to find.
With the help of next‐generation sequencing technology, genetic abnormalities in NOTCH1 have been implicated in leukemia and solid tumors (Liu et al., 2016; Malecki et al., 2006; Riley, McBride, & Cole, 2011; Song et al., 2014). Most of the NOTCH1 mutations in hematopoietic tumors are clustered in the transmembrane and intracellular domains. Patients with these mutations are often suggested poor prognosis (Aster et al., 2000; Palomero, Dominguez, & Ferrando, 2008). Michael J et al. have reported that leukemia‐associated HD domain mutations can activate Notch1 independent of stimulation by canonical ligands. Notch1 receptors bearing mutations can reach the cell surface and release sufficient
NICD to activate Notch1 signaling (Malecki et al., 2006). Previously, our group shows that in Chinese patients with OSCC, the NOTCH1 mutation frequency is greater than 40% and strongly associated with poor prognosis and shorter survival (Song et al., 2014). Moreover, the spectrum of NOTCH1 mutations we observe in OSCC is mostly in the N‐terminal epidermal growth factor (EGF) within NECD and hints to be Notch1‐inactivating mutations accompanied by defective ligand binding, which is fundamentally different from those in hematopoietic tumors (Agrawal et al., 2011; Song et al., 2014).
Further studies have confirmed our dissertation. Patients with Notch1 somatic mutations in EGF‐like domain have significantly higher recurrence rate in either 5‐ or 10‐year period and a lower survival rate (Liu et al., 2016). The saltation is largely located to EGF‐like domain that may functionally compromise and increase OSCC recurrence and fatality, suggesting that Notch1 performs an indispensable role in OSCC (Liu et al., 2016; Rettig et al., 2015).
Despite Notch1 has been reported to be mutated at a significant frequency in OSCC, the possible role and the consequence of the regulation of Notch1 signaling remain controversial (Inamura et al., 2017; Kwon et al., 2011; Purow et al., 2008; Weijzen et al., 2002). It has been previously found that the intracellular cleavage (S3) of Notch1 occurs between the amino acids Gly1753 and Val1754 in human in a highly conserved manner (Schroeter, Kisslinger, & Kopan, 1998; Figure 1a). Mutations of Val1754 (V1754K or V1754L) in human impede the intracellular cleavage of Notch1, leaving the Notch1 protein tethered to the membrane (membrane‐tethered Notch1) and thereby reducing the downstream activities of Notch1 (Kopan, Schroeter, Weintraub, & Nye, 1996; Schroeter et al., 1998). To imitate the inactivation of Notch1 due to inappropriate mutations and to determine the potential mechanism of membrane‐tethered Notch1, we utilized wild‐type Notch1 vectors containing full‐length wild‐type Notch1 complementary DNA (cDNA; Notch1WT) or vectors containing mutant Notch1 cDNA (Notch1V1754L) to transfect into OSCC cell lines. Oncogenic phenotypes were discovered in both transfected cell lines. In particular, associated with the deficiency of canonical Notch1 signal transcription, epidermal growth factor receptor (EGFR)–phosphoinositide 3‐kinase (PI3K)–protein kinase B (AKT) pathway has been distinctively activated by membrane‐tethered Notch1. This finding may imply the significance of noncanonical Notch1 pathway in the absence of conventional Notch1 activation, which has not been thoroughly discussed before. Thus, the compensation of EGFR–PI3K– AKT pathway may provide a crucial approach for controlling OSCC with targetable Notch1 mutations.


2.1 | Cell culture and inhibitor

The human OSCC cell lines CAL27 and SCC9 were purchased from the American Type Culture Collection (ATCC, 10801 University Notch1 in stable CAL27 cells after transfection. Notch1WT/V1754L‐transfected CAL27 cells exhibited much stronger Notch1 immunofluorescence staining than the control cells. In Notch1WT‐transfected cells, Notch1 staining was distributed in the cytoplasm or membrane as well as in the nucleus, whereas the Notch1V1754L‐transfected cells mainly exhibited higher levels of membranous and cytoplasmic Notch1 staining, but not nuclear staining. In the process of analyzing, 100 cells in five different sites were assessed and the MN rate (MN vs. DAPI ratio) was shown as mean ± SD. MN in Notch1V1754L‐ transfected cells (67 ± 5%) was much higher than that in Notch1WT‐transfected cells (21 ± 4%). (e) Real‐time qPCR was utilized to determine the expression of Notch1 downstream target genes HES‐1 and HES‐2 in the cells (SCC9 and CAL27) after transfection. (f) Notch1 signaling downstream target genes were evaluated by western blot Boulevard, Manassas, VA). Cells were cultured in Dulbecco’s modified Eagle’s medium and Ham’s F12 medium supplemented with 10% fetal bovine serum and 100 U/ml penicillin–streptomycin (Invitrogen, 3175 Staley Road, Grand Island, NY) in humidified incubators at 37°C with 5% CO2. γ‐Secretase inhibitor PF‐03084014 was purchased from Selleck Chem, 9330 Kirby Drive, STE 200, Houston, TX and was dissolved in dimethyl sulphoxide which was also used for treatment control.

2.2 | Plasmid construction and transfection experiments

Wild‐type Notch1 vectors containing full‐length wild‐type Notch1 cDNA (Notch1WT) or mutant Notch1 vectors containing mutant Notch1 cDNA (Notch1V1754L) inserted into Flag‐pcDNA 3.1 vectors were synthesized and constructed by Generay Biotech (Shanghai, China). For transfection, cells (5 × 105 per well in six‐well plates) were cultured to 50% confluence in complete growth medium, after which the medium was replaced with serum‐free medium for 4–6 hr. Purified plasmids were transfected into the cells using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. Forty‐eight hours after transfection, the medium was supplemented with 200–400 μg/ml G418 (Invitrogen). Two to three weeks later, cells selected from monoclones were subcultured and harvested. Transfection efficiency was evaluated by real‐time quantitative polymerase chain reaction (qPCR) and western blot analysis using anti‐Notch1 antibodies (see below). Cells from three monoclones were set as repeated trials.

2.3 | Western blot analysis

Total protein was lysed for 30 min using lysis buffer (Beyotime, Shanghai, China) containing protease inhibitor cocktail and phosphatase inhibitor cocktail (Selleck Chem, 9330 Kirby Drive, STE 200, Houston, TX). Coomassie Brilliant Blue was used to quantify the protein lysates, and bovine serum albumin was used as the standard. The proteins (10 μg, 2 μg/μl) were separated using sodium dodecyl sulfate‐polyacrylamide gels and transferred onto polyvinylidene fluoride membranes (Millipore, Munich, Germany), and were then blocked with 5% skimmed milk at room temperature for 2 hr and hybridized with primary antibodies (diluted 1:1,000) specific for Notch1 (clone: D1E11), NICD (or cleaved NICD; clone: D3B8), cyclin‐dependent kinase (CDK) 2/4, cyclin D3, P21, EGFR, phospho‐EGFR (p‐EGFR), AKT, p‐AKT (Cell Signaling Technology, 3 Trask Lane, Danvers, MA), PI3K, E‐cadherin, N‐cadherin, vimentin, β‐catenin, β‐actin or Flag‐tag (Bioworld, Nanjing, China) overnight at 4°C, followed by incubation with anti‐goat immunoglobulin G (IgG) horseradish peroxidase (HRP)–conjugated second antibodies (Zhongshan Golden Bridge Bio, Beijing, China) for 1 hr at room temperature. The immunoreactive bands were detected using an Immobilon Western Chemiluminescent HRP Substrate (Milli- pore) and visualized using the ImageQuant LAS4000 Mini Imaging System (General Electrics, Louisville, Kentucky). Analyses of the bands were performed with ImageJ, Rawak Software Inc., Germany.

2.4 | Real‐time qPCR

Total RNA was extracted from the cells using TRIzol reagent (Invitrogen) and was converted to cDNA using 5× PrimeScript RT Master Mix (TaKaRa, Otsu, Japan) at 37°C for 15 min and 85°C for 5 s according to the manufacturer’s protocol. Obtained cDNA (1 μg/well) was used to perform real‐time reverse‐transcription PCR by SYBR Premix Ex Taq Kit (TaKaRa) and ABI 7900 real‐time PCR system (Applied Biosystems, 3175 Staley Road, Grand Island, NY). The ΔΔCt method was used for the quantitation of relative gene expression to determine the mean expression of each target gene normalized to the geometric mean of GAPDH. All primers were designed and synthesized to target specific sequences of the genes as follows:

2.5 | Immunofluorescence staining

CAL27 cells (2× 104) stably expressing Flag‐pcDNA3.1‐Notch1WT, Flag‐ pcDNA3.1‐Notch1V1754L, or Flag‐pcDNA3.1 vector were seeded onto sterile glass coverslips in 24‐well plates at approximately 20% confluence. After treatment for 24 hr, cells were fixed in 4% paraformaldehyde and permeabilized in 1% Triton. After incubation overnight with rabbit anti‐Notch1 antibody (1:100; D1E11; Cell Signaling Technology), cells were stained for 1 hr with goat anti‐rabbit IgG antibody Cy3 (Abcam, 330 Cambridge Science Park, Cambridge, CB4 0FL, UK) and counterstained with 4′,6‐diamidino‐2‐phenylindole (DAPI; Sigma‐Aldrich, St. Louis, MO). Cells were subsequently viewed by a fluorescence microscopy (ZEISS, Oberkochen, Germany).

2.6 | Transwell invasion assays

According to the manufacturer’s instructions, transwell chambers (8‐μm pore size; Millipore) were used to detect cell invasive ability. A total of 3 × 104 cells were seeded into the upper chamber of each insert and incubated at 37°C for 24 hr. Similar inserts coated with Matrigel (BD Biosciences, 2350 Qume Drive San Jose, CA) were used to determine invasive potential in cell invasion assays. Chambers were fixed in 4% paraformaldehyde for 30 min and then dyed with crystal violet staining. The noninvaded cells on the upper chamber surface were removed, and the invaded cells on the surface were subsequently viewed and counted by a microscopy (ZEISS).

2.7 | Cell Counting Kit‐8 (CCK‐8) assays

Cells were seeded into 96‐well microplates at a density of 2 × 103 cells per well and incubated in fresh medium containing 10% CCK‐8 reaction solution (Dojindo, Tokyo, Japan). After incubation for 1 hr, the absorbance was measured on a microplate spectro- photometer (Multiskan MK3; Thermo Fisher Scientific, 3175 Staley Road, Grand Island, NY) at a wavelength of 450 nm according to the manufacturer’s instructions. Five independent experiments were performed. The growth curves were illustrated using GraphPad Prism 6 software, 7825 Fay Avenue, Suite 230, La Jolla, CA.

2.8 | Flow cytometry

For cell‐cycle analysis, stable cells were harvested and washed in phosphate‐buffered saline (PBS) and fixed in 75% ice‐cold ethanol for 30 min at 4°C. Cells were then washed twice in PBS, stained with propidium iodide (50 μg/ml) in the presence of 50 μg/ml RNase A (Sigma‐Aldrich) and incubated for 1 hr at room temperature. The cell‐ cycle analysis was performed on a FACSCalibur flow cytometer BD Biosciences) and CellQuest Pro software (BD Biosciences). Flow cytometric analysis of apoptotic cells was performed by staining the cells using the Annexin V Apoptosis Detection Kit (BD Pharmingen, 2350 Qume Drive San Jose, CA) according to the manufacturer’s protocol. The percentages of cells in specific cell‐cycle stages in Notch1V1754L cells were compared with those in Notch1WT cells.

2.9 | Statistical analyses

Statistical analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, IL). All experiments were repeated at least three times. Each data point represents the mean ± SD of data from three independent trials. The Student’s t test was used to evaluate the statistical significance of the results. All p values represented two‐sided tests of statistical significance (*p < 0.05, **p < 0.01, and ***p < 0.001). 3 | RESULTS 3.1 | Notch1 mutant transfection inactivated canonical Notch1 signaling To mimic the membrane‐tethered Notch1, we mutated the Val1754 residue of the human Notch1, which is equivalent to Val1744 in the mouse (Huppert et al., 2000), to assess its association with Notch1 functions in OSCC cell lines (Figure 1a). SCC9 and CAL27 cells were transfected with vectors containing wild‐type Notch1 (Flag‐pcDNA3.1‐Notch1WT), Notch1 with Val1754L mutation (Flag‐pcDNA3.1‐Notch1V1754L), or empty Flag‐pcDNA3.1 vector (hereafter referred to as WT, 1754, or P, respectively). After selection, stable SCC9–CAL27‐Notch1V1754L/WT or ‐P cells were obtained, and the transfection efficiency was verified by real‐time qPCR using primers not covering the mutant nucleotide (Figure 1b), or western blot analysis by detecting the Notch1 expression and the S3 cleavage activity (Figure 1c). The membrane‐tethered Notch1 induced by the mutation in Val1754 was also detected by immunofluorescence (Figure 1d) using the Notch1‐specific antibody. As shown, we found that the Notch1WT‐transfected CAL27 cells exhibited much stronger Notch1 immunofluorescence signals than the control cells and that Notch1 was distributed in the cytoplasm–membrane as well as in the nucleus, whereas the Notch1V1754L‐transfected cells mainly exhibited higher levels of membrane‐tethered and cytoplasmic Notch1 staining, but not nuclear staining. We then focused on the membrane‐tethered Notch1 expression pattern. The quantitative percentage of membranous Notch1 rate (MN vs. DAPI ratio) in Notch1V1754L‐transfected cells was much higher than that in Notch1WT‐transfected cells (p < 0.01; Figure 1d). Considering the impaired nuclear translocation of Notch1, the membrane‐tethered Notch1 induced by Notch1V1754L transfection could be functionally confirmed. To verify whether the membrane‐tethered Notch1 influenced the Notch1 canonical pathway, we examined Notch1 signaling pathway activity. We performed real‐time qPCR to determine the expression of HES genes in the cells (SCC9 and CAL27) after transfection. The results showed that compared to the cells transfected with vectors containing wild‐type Notch1, the cells transfected with vectors containing the V1754L mutation in Notch1 could drastically reduce HES‐1 and HES‐2 messenger RNA (mRNA) and protein expression (Figure 1e,f). These results strongly suggested that the V1754L mutation in Notch1 inactivated Notch1 signaling. 3.2 | Membrane‐tethered Notch1 enhanced invasion and induced epithelial‐to‐mesenchymal transition (EMT) in OSCC cell lines To determine whether membrane‐tethered Notch1 altered invasion of OSCC cells, we utilized Transwell invasion assays. As shown in Figure 2a, the Notch1WT‐transfected cells exhibited drastically increased invasive ability compared to the control cells. In addition, our results revealed that Notch1V1754L‐transfected cells exhibited even higher invasive ability in both cell lines compared with Notch1WT‐transfected cells. During malignant progression, it has been proposed that carcinoma cells undergo an EMT, in which they lose epithelial characteristics and acquire invasive properties (Katafiasz, Smith, & Wahl, 2011; Thiery, Acloque, Huang, & Nieto, 2009; Yang et al., 2008). It is characterized by the downregulation of epithelial markers (e.g., β‐catenin and E‐cadherin) and the upregulation of mesenchymal markers (e.g., N‐cadherin and vimentin; Thompson, Newgreen, & Tarin, 2005). In addition, Notch1 plays as a fundamental regulator in the induction as well as the maintenance of EMT and tumor progression (Koch, Lehal, & Radtke, 2013; Yuan et al., 2014). During the procedure of culturing, we noticed that compared to the control cells, the Notch1WT‐transfected cells adopted a markedly altered cellular morphology characterized by a spindle shape and presence of pseudopodia (Figure 2b), suggesting the loss of cell‐cell adhesion; the Notch1V1754L mutation did not recover the original cobblestone‐like epithelial morphology. These changes implied that overexpression of Notch1 induced an EMT‐like phenotype and that the loss‐of‐function of Notch1 by V1754L mutation did not reverse the trend. Additionally, we performed western blot analysis to assess the expression of EMT‐specific markers at the protein level. As expected, in CAL27 cells, the transfection of Notch1WT vector slightly increased the expression of the mesenchymal marker vimentin and drastically reduced the expression of the epithelial marker β‐catenin (or slightly reduced E‐cadherin) compared to the control cells (Figure 2c). Interestingly, the mesenchymal marker N‐cadherin was slightly downregulated by Notch1WT, not upregu- lated as anticipated. Meanwhile, Notch1V1754L‐transfected cells exhibited a marked increase in the expression of vimentin, and a slight increase in the expression of N‐cadherin and β‐catenin, but showed decreased expression of E‐cadherin compared to the Notch1WT‐transfected cells (Figure 2c). 3.3 | Membrane‐tethered Notch1 increased proliferation in OSCC CCK‐8 assays were performed to test the viability of the OSCC cells. Data demonstrated that the Notch1WT‐transfected cells had higher optical density (OD) values than the control cells; the Notch1V1754L‐ transfected cells had even higher or maintained high OD values compared with the Notch1WT‐transfected cells (Figure 3a,b). Moreover, we performed cell‐cycle analysis by flow cytometry. The results demonstrated that Notch1WT‐ and Notch1V1754L‐transfected cells presented a significantly lower percentage of them in the G1 phase than control cells. Meanwhile, the proportion of Notch1V1754L‐transfected cells which were in the G1 phase was even lower than that of Notch1WT‐transfected cells in the same phase (Figure 3c). In T‐cell lymphomas, context‐specific putative target genes have been identified through which Notch1 may promote transformation by altering cell‐growth kinetics (Weng et al., 2006). The D‐type cyclins (including cyclin D3) are reported to be induced as cells enter the G1 phase of the cell cycle and can activate CDK, CDK4, and CDK6, which can drive the cell into the S phase (Joshi et al., 2009). P21 as an inhibitor of CDK2 can decrease cell proliferation through blocking the G1 and the S phase in cell cycle (Labaer et al., 1997). In this study, changes in the expression of cell‐cycle‐specific proteins were selectively analyzed by western blot analysis. Consistently, Notch1V1754L transfection showed the upregulation of CDKs (2 and 4) and cyclins (D3), as well as the downregulation of P21 (Figure 3d). These results suggested that membrane‐tethered Notch1 enhanced the proliferative ability of the cells by accelerating cell cycle. 3.4 | Membrane‐tethered Notch1 promoted tumor progression by modulating EGFR–PI3K–AKT signaling pathway EGFR has an important role in cell proliferation, differentiation, and transformation (Chiang et al., 2008). Several alterations and activation of EGFR have been identified in tumors. EGFR–PI3K–AKT signal has been linked with oncogenic transformation, autonomous cell growth, invasion, angiogenesis, and development of metastases in oral cancers (Sos et al., 2015). Notch1 has been shown to interact with EGFR–PI3K–AKT pathway and thus to regulate cell proliferation and invasion (Kolev et al., 2012; Staberg et al., 2016; Zheng et al., 2017). Therefore, we tested the Notch1‐mediated regulation of EGFR– PI3K–AKT signaling pathway in CAL27 and SCC9 cells. We found that the levels of EGFR and p‐EGFR were reduced after Notch1WT transfection, whereas the levels of AKT, p‐AKT, and PI3K were steadily increased, suggesting the activated status of the PI3K–AKT signaling pathway. Meanwhile, the Notch1V1754L‐transfected cells exhibited remarkable upregulated p‐EGFR expression. Cells sustained expression of PI3K–AKT proteins compared to the Notch1WT‐ transfected cells, suggesting the presence of activated EGFR– PI3K–AKT pathway (Figure 4a). Notably, EGFR regulation did not occur at transcriptional level, because mRNA abundance was not significantly affected (Figure 4b). 3.5 | γ‐Secretase inhibitor reduced the NICD translocation and enhanced EGFR–PI3K–AKT pathway activation Before NICD is imported into nucleus, Notch1 receptor has to undergo a series of programmed proteolytic events, first by α‐secretase at the extracellular surface, which leads to liberation of the extracellular fragment, and then by intramembranous cleavage mediated by γ‐secretase (Shih Ie & Wang, 2007). The effectiveness of γ‐secretase ensures proteolysis‐dependent activation of Notch1 and thus initiates canonical Notch1 pathway. As a γ‐secretase inhibitor, PF‐03084014 predominantly prevents the generation of intracellular domain of Notch1 molecules and suppresses the Notch1 activity (Lanz et al., 2010; Wei et al., 2010). To verify the interaction between noncanonical Notch1 and EGFR–PI3K–AKT signaling, PF‐03084014 was experimented in OSCC cells. As SCC9 or CAL27 cells was treated with 10 μM PF‐ 03084014, NICD was almost undetectable by western blot analysis (Figure 5a). Although it is acknowledged that PF‐03084014 functions to induce cell growth arrest and exhibits antitumoral and antiangiogenic efficacy in tumors (Cathy et al., 2014; Samon et al., 2012), we noticed an enhancement in cell proliferation and invasion in CAL27 cells (Figure 5b–d). In addition, the expression of p‐EGFR, p‐AKT, and PI3K prominently increased after PF‐03084014 treatment (Figure 5e). Consistent with the result of membrane‐tethered Notch1 transfection, the accumulation of inactivated Notch1 protein obtained compensation from another pathway to exhibit oncogenic property. 4 | DISCUSSION The NOTCH1 gene encodes a single‐pass transmembrane receptor that is responsible for cell fate determination during several steps of metazoan development (Huppert et al., 2000). Its roles in malignant initiation and progression, however, are complex and appear cell or tissue dependent. In contrast with the oncogenic roles of Notch1 observed in leukemia, tumor‐suppressive roles of the Notch1 pathway have been recently noted in multiple tumor types (Mao, 2015). Although subsequent studies have confirmed the NOTCH1 inactivating mutations by whole‐genome sequencing methods and demonstrated Notch1 tumor‐suppressor activity in OSCC cell lines, Notch1 dysregulation appears to be more complex than simple loss‐ of‐function (Rettig et al., 2015). It is well recognized that ligand binding to Notch1 triggers a proteolytic release from the membrane, resulting in amounts of NICD which form a nuclear complex to activate downstream targets. To date, no observations have clearly simulated the Notch1 activities by blocking its processing and then determine the ability of unprocessed Notch1 to signal. Here we mimicked membrane‐tethered Notch1 expression and reported the oncogenic phenotype associated with a single‐point mutation at the intramembranous processing site of Notch1, V1754L. In our study, we introduced the V1754L mutation in Notch1 in an overexpression cellular model. This cellular model confirmed an inactivation of canonical Notch1 pathway with a reduction of activated NICD and downstream targets compared with wild‐type Notch1, although mere membranous Notch1 could not be detected by immunofluorescence due to the transfection‐induced persistent translation of Notch1 protein in the cytoplasm. Interestingly, we noticed that in SCC9 or CAL27 cells transfected with Notch1V1754L, NICD levels were slightly elevated compared with Notch1P (Figure 1c) cells. We assumed that the accumulated binding of mutated Notch1 may result in the overexpression of ligands presenting at the neighboring cells, which in turn causes endogenous Notch1 transactivation(Miller, Lyons & Herman, 2009; Del Álamo, Rouault, & Schweisguth, 2011). As HES‐1 and HES‐2 (downstream targets of Notch1) protein levels remained unchanged (Figure 1f), further experiments need to be performed to verify the hypothesis. EMT is a fundamental process governing morphogenesis, characterized by the loss of E‐cadherin and a concurrent increase in N‐cadherin and/or vimentin expression. Previous in vitro studies in breast cancer cell lines revealed a possible mechanism of Notch1‐ induced EMT, in which Notch1, in complex with the transcription factor CSL, subsequently decreases E‐cadherin expression and results in increased invasiveness (Dubois‐Marshall, Thomas, Faratian, Harrison, & Katz, 2011). In this study, we verified the enhancement of invasiveness after Notch1 transfection by inducing EMT markers. Meanwhile, we found that in mutated Notch1‐transfected cells, vimentin played a significant role in mesenchymal transition. This result implied that the transfection of wild‐type Notch1 induces a Notch1‐mediated and ligand‐dependent oncogenic phenotype, while the retention of the membrane‐tethered Notch1 protein mediates tumor progression via other potential mechanisms, independent of intracellular cleavage. As a transmembrane receptor for members of the EGF family, EGFR has been implicated in the pathogenesis of various tumors. It is found to be upregulated and overexpressed in the majority of oral cancers and is associated with a poor clinical prognosis (Chiang et al., 2008). Activation of PI3K–AKT signaling through activation of EGFR signaling molecules occurs in a mass of cancers contributing to deregulation of proliferation as well as resistance to apoptosis (O’Reilly et al., 2006). Accumulating evidence supports the crosstalk between the Notch1 pathway and others, such as the EGFR pathway (Aguirre, Rubio, & Gallo, 2010; Fassl et al., 2012), which contributes to the complexity of Notch1 signaling in development and disease (Guruharsha, Kankel, & Artavanis‐Tsakonas, 2012). Crosstalk between the EGFR and Notch1 pathways has been identified in breast cancer (Baker, Zlobin, & Osipo, 2014), lung cancer (Konishi et al., 2010), skin cancer (Kolev et al., 2008), and gliomas (Purow et al., 2008). The two pathways can function in either an antagonistic or synergistic manner, depending on the tissue and developmental context (Kolev et al., 2008). We verified this crosstalk and found that in OSCC cell lines, although wild‐type Notch1 downregulated p‐EGFR, it upregulated downstream p‐AKT and PI3K, suggesting the activated downstream pathway. Addition- ally, the membrane‐tethered Notch1 resulting from V1754L mutation maintained or even enhanced the activated status of the EGFR–PI3K–AKT signaling pathway, which was consistent with its oncogenic phenotype. Moreover, EGFR mRNA levels did not show a statistical discrepancy. To determine the endogenous Notch1 activity, we further treated cells with Notch1 pathway inhibitor, PF‐03084014. Compared with control, PF‐03084014 exhibited a reduction of cleaved NICD, thus verified the block of pathway transduction. Moreover, although PF‐03084014 has been reported to exert its antitumor activities through direct effects on tumor cell cycle and apoptosis mechanisms (Wei et al., 2010), our study discovered distinct oncogenic property after treatment. Inhibition of canonical Notch1 pathway resulted in enhanced cell proliferation and invasion, indicating its paradoxical role in OSCC. We then found an abnormal activation in EGFR–PI3K–AKT pathway after PF‐03084014 treatment. The interplay between two pathways in this experiment may elucidate the functional controversy in Notch1 signaling. Above all, this was the first time that we clearly simulated the Notch1 activities by blocking its processing and determined the ability of unprocessed Notch1 to signal. These data emphasized the significant role of EGFR–PI3K–AKT signaling in Notch1‐mediated oncogenesis and could explain the potential mechanism underlying membrane‐tethered Notch1‐related phenotypic alterations. Based on the increasing attention paid to the function of membrane‐tethered Notch1 in the noncanonical Notch1 signaling pathway, further study is required to explore the exact role of membrane‐tethered Notch1 in OSCC. REFERENCES Agrawal, N., Frederick, M. J., Pickering, C. R., Bettegowda, C., Chang, K., Li, R. J., Myers, J. N. (2011). Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1. Science, 333(6046), 1154–1157. Aguirre, A., Rubio, M. E., & Gallo, V. (2010). Notch and EGFR pathway interaction regulates neural stem cell number and self‐renewal. Nature, 467(7313), 323–327. Aster, J. C., Xu, L., Karnell, F. G., Patriub, V., Pui, J. C., & Pear, W. S. (2000). Essential roles for ankyrin repeat and transactivation domains in induction of T‐cell leukemia by notch1. Molecular and Cellular Biology, 20(20), 7505–7515. Baker, A. T., Zlobin, A., & Osipo, C. (2014). Notch‐EGFR/HER2 bidirectional crosstalk in breast cancer. Frontiers in Oncology, 4, 360. Cathy, C., Zhang, Z. Y., Giddabasappa, A., Lappin, P. B., Painter, C. L., Zhang, Q., Xie, Z. (2014). Comparison of dynamic contrast‐ enhanced MR, ultrasound and optical imaging modalities to evaluate the antiangiogenic effect of PF‐03084014 and sunitinib. Cancer Medicine, 3(3), 462–471. Chiang, W. F., Liu, S. Y., Yen, C. Y., Lin, C. N., Chen, Y. C., Lin, S. C., & Chang, K. W. (2008). Association of epidermal growth factor receptor (EGFR) gene copy number amplification with neck lymph node metastasis in areca‐associated oral carcinomas. Oral Oncology, 44(3), 270–276. De Strooper, B., Annaert, W., Cupers, P., Saftig, P., Craessaerts, K., Mumm, J. S., Kopan, R. (1999). A presenilin‐1‐dependent gamma‐secretase‐ like protease mediates release of Notch intracellular domain. Nature, 398(6727), 518–522. Del Álamo, D., Rouault, H., & Schweisguth, F. (2011). Mechanism and significance of cis‐inhibition in Notch signalling. Current Biology, 21(1), R40–R47. Dubois‐Marshall, S., Thomas, J. S., Faratian, D., Harrison, D. J., & Katz, E. (2011). Two possible mechanisms of epithelial to mesenchymal transition in invasive ductal breast cancer. Clinical and Experimental Metastasis, 28(8), 811–818. Fassl, A., Tagscherer, K. E., Richter, J., Berriel Diaz, M., Alcantara Llaguno, S. R., Campos, B., Roth, W. (2012). Notch1 signaling promotes survival of glioblastoma cells via EGFR‐mediated induction of anti‐ apoptotic Mcl‐1. Oncogene, 31(44), 4698–4708. Guruharsha, K. G., Kankel, M. W., & Artavanis‐Tsakonas, S. (2012). The Notch signalling system: Recent insights into the complexity of a conserved pathway. Nature Reviews Genetics, 13(9), 654–666. Huppert, S. S., Le, A., Schroeter, E. H., Mumm, J. S., Saxena, M. T., Milner, L. A., & Kopan, R. (2000). Embryonic lethality in mice homozygous for a processing‐deficient allele of Notch1. Nature, 405(6789), 966–970. Inamura, N., Kimura, T., Wang, L., Yanagi, H., Tsuda, M., Tanino, M., Tanaka, S. (2017). Notch1 regulates invasion and metastasis of head and neck squamous cell carcinoma by inducing EMT through c‐Myc. Auris, Nasus, Larynx, 44(4), 447–457. Izumchenko, E., Sun, K., Jones, S., Brait, M., Agrawal, N., Koch, W., Sidransky, D. (2015). Notch1 mutations are drivers of oral tumorigen- esis. Cancer Prevention Research, 8(4), 277–286. Jeffries, S., & Capobianco, A. J. (2000). Neoplastic transformation by Notch requires nuclear localization. Molecular and Cellular Biology, 20(11), 3928–3941. Joshi, I., Minter, L. M., Telfer, J., Demarest, R. M., Capobianco, A. J., Aster, J. C., Osborne, B. A. (2009). Notch signaling mediates G1/S cell‐ cycle progression in T cells via cyclin D3 and its dependent kinases. Blood, 113(8), 1689–1698. Katafiasz, D., Smith, L. M., & Wahl, J. K., 3rd (2011). Slug (SNAI2) expression in oral SCC cells results in altered cell‐cell adhesion and increased motility. Cell Adhesion and Migration, 5(4), 315–322. Koch, U., Lehal, R., & Radtke, F. (2013). Stem cells living with a Notch. Development, 140(4), 689–704. Kolev, V., Mandinova, A., Guinea‐Viniegra, J., Hu, B., Lefort, K., Lambertini, C., Dotto, G. P. (2008). EGFR signalling as a negative regulator of Notch1 gene transcription and function in proliferating keratinocytes and cancer. Nature Cell Biology, 10(8), 902–911. Konishi, J., Yi, F., Chen, X., Vo, H., Carbone, D. P., & Dang, T. P. (2010). Notch3 cooperates with the EGFR pathway to modulate apoptosis through the induction of bim. Oncogene, 29(4), 589–596. Kopan, R., Schroeter, E. H., Weintraub, H., & Nye, J. S. (1996). Signal transduction by activated mNotch: Importance of proteolytic proces- sing and its regulation by the extracellular domain. Proceedings of the National Academy of Sciences of the United States of America, 93(4), 1683–1688. Kwon, C., Cheng, P., King, I. N., Andersen, P., Shenje, L., Nigam, V.,& Srivastava, D. (2011). Notch post‐translationally regulates beta‐catenin protein in stem and progenitor cells. Nature Cell Biology, 13(10), 1244–1251. Labaer, J., Garrett, M. D., Stevenson, L. F., Slingerland, J. M., Sandhu, C., Chou, H. S., Harlow, E. (1997). New functional activities for the p21 family of CDK inhibitors. Genes and Development, 11(7), 847–862. Lanz, T. A., Wood, K. M., Richter, K. E., Nolan, C. E., Becker, S. L., Pozdnyakov, N., Tate, B. (2010). Pharmacodynamics and pharma- cokinetics of the gamma‐secretase inhibitor PF‐3084014. Journal of Pharmacology and Experimental Therapeutics, 334(1), 269. Le Borgne, R. (2006). Regulation of Notch signalling by endocytosis and endosomal sorting. Current Opinion in Cell Biology, 18(2), 213–222. Liu, Y. F., Chiang, S. L., Lin, C. Y., Chang, J. G., Chung, C. M., Ko, A. M. S., Ko, Y. C. (2016). Somatic mutations and genetic variants of NOTCH1 in head and neck squamous cell carcinoma occurrence and develop- ment. Scientific Reports, 6, 24014. Ma, J., Tang, X., Wong, P., Jacobs, B., Borden, E. C., & Bedogni, B. (2014). Noncanonical activation of Notch1 protein by membrane type 1 matrix metalloproteinase (MT1‐MMP) controls melanoma cell proliferation. The Journal of Biological Chemistry, 289(12), 8442–8449. Malecki, M. J., Sanchez‐Irizarry, C., Mitchell, J. L., Histen, G., Xu, M. L., Aster, J. C., & Blacklow, S. C. (2006). Leukemia‐associated mutations within the NOTCH1 heterodimerization domain fall into PF-03084014 at least two distinct mechanistic classes. Molecular and Cellular Biology, 26(12), 4642–4651.
Mao, L. (2015). NOTCH mutations: Multiple faces in human malignancies. Cancer Prevention Research, 8(4), 259–261.
Mardis, E. R. (2008). The impact of next‐generation sequencing technology on genetics. Trends in Genetics, 24(3), 133–141.
McGill, M. A., & McGlade, C. J. (2003). Mammalian numb proteins promote Notch1 receptor ubiquitination and degradation of the Notch1 intracellular domain. The Journal of Biological Chemistry, 278(25), 23196–23203.
Miller, A. C., Lyons, E. L., & Herman, T. G. (2009). Cis‐inhibition of Notch by endogenous Delta biases the outcome of lateral inhibition. Current Biology, 19(16), 1378–1383.
O’Reilly, K. E., Rojo, F., She, Q. B., Solit, D., Mills, G. B., Smith, D., Rosen, N. (2006). MTOR inhibition induces upstream receptor tyrosine kinase signaling and activates Akt. Cancer Research, 66(3), 1500–1508. Palomero, T., Dominguez, M., & Ferrando, A. A. (2008). The role of the PTEN/AKT pathway in NOTCH1‐induced leukemia. Cell Cycle, 7(8), 965–970.
Palomero, T., Sulis, M. L., Cortina, M., Real, P. J., Barnes, K., Ciofani, M., Ferrando, A. A. (2007). Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T‐cell leukemia. Nature Medicine, 13(10), 1203–1210.
Pei, Z., & Baker, N. E. (2008). Competition between Delta and the Abruptex domain of Notch. BMC Developmental Biology, 8, 4.
Purow, B. W., Sundaresan, T. K., Burdick, M. J., Kefas, B. A., Comeau, L. D., Hawkinson, M. P., Fine, H. A. (2008). Notch‐1 regulates transcrip- tion of the epidermal growth factor receptor through p53. Carcino- genesis, 29(5), 918–925.
Rettig, E. M., Chung, C. H., Bishop, J. A., Howard, J. D., Sharma, R., Li, R. J., Fakhry, C. (2015). Cleaved NOTCH1 expression pattern in head and neck squamous cell carcinoma is associated with NOTCH1 mutation, HPV status and high‐risk features. Cancer Prevention Research, 8(4), 287–295.
Riley, M. F., McBride, K. L., & Cole, S. E. (2011). NOTCH1 missense alleles associated with left ventricular outflow tract defects exhibit impaired receptor processing and defective EMT. Biochimica et Biophysica Acta, 1812(1), 121–129.
Samon, J. B., Castillo‐Martin, M., Hadler, M., Ambesi‐Impiobato, A., Paietta, E., Racevskis, J., Ferrando, A. A. (2012). Preclinical analysis of the gamma‐secretase inhibitor PF‐03084014 in combination with glucocorticoids in T‐cell acute lymphoblastic leukemia. Molecular Cancer Therapeutics, 11(7), 1565–1575.
Schroeter, E. H., Kisslinger, J. A., & Kopan, R. (1998). Notch‐1 signalling requires ligand‐induced proteolytic release of intracellular domain. Nature, 393(6683), 382–386.
Sharma, A., Rangarajan, A., & Dighe, R. R. (2013). Antibodies against the extracellular domain of human Notch1 receptor reveal the critical role of epidermal‐growth‐factor‐like repeats 25‐26 in ligand binding and receptor activation. The Biochemical Journal, 449(2), 519–530.
Shih Ie, M., & Wang, T. L. (2007). Notch signaling, gamma‐secretase inhibitors, and cancer therapy. Cancer Research, 67(5), 1879–1882.
Song, X., Xia, R., Li, J., Long, Z., Ren, H., Chen, W., & Mao, L. (2014). Common and complex Notch1 mutations in Chinese oral squamous cell carcinoma. Clinical Cancer Research, 20(3), 701–710.
Sos, M. L., Koker, M., Weir, B. A., Heynck, S., Rabinovsky, R., Zander, T., Frommolt, P. (2015). PTEN loss contributes to erlotinib resistance in EGFR‐mutant lung cancer by activation of Akt and EGFR. Cancer Research, 75(9), 3256.
Staberg, M., Michaelsen, S. R., Olsen, L. S., Nedergaard, M. K., Villingshøj, M., Stockhausen, M. T., Poulsen, H. S. (2016). Combined EGFR‐ and notch inhibition display additive inhibitory effect on glioblastoma cell viability and glioblastoma‐induced endothelial cell sprouting in vitro. Cancer Cell International, 16, 34.
Stransky, N., Egloff, A. M., Tward, A. D., Kostic, A. D., Cibulskis, K., Sivachenko, A., Grandis, J. R. (2011). The mutational landscape of head and neck squamous cell carcinoma. Science, 333(6046), 1157–1160.
Sun, W., Gaykalova, D. A., Ochs, M. F., Mambo, E., Arnaoutakis, D., Liu, Y., Califano, J. A. (2014). Activation of the NOTCH pathway in head and neck cancer. Cancer Research, 74(4), 1091–1104.
Thiery, J. P., Acloque, H., Huang, R. Y. J., & Nieto, M. A. (2009). Epithelial‐ mesenchymal transitions in development and disease. Cell, 139(5), 871–890.
Thompson, E. W., Newgreen, D. F., & Tarin, D. (2005). Carcinoma invasion and metastasis: A role for epithelial‐mesenchymal transition. Cancer Research, 65(14), 5991–5995.
Traustadóttir, G. Á., Jensen, C. H., Thomassen, M., Beck, H. C., Mortensen, S. B., Laborda, J., Andersen, D. C. (2016). Evidence of non‐canonical NOTCH signaling: Delta‐like 1 homolog (DLK1) directly interacts with the NOTCH1 receptor in mammals. Cellular Signalling, 28(4), 246–254.
Wei, P., Walls, M., Qiu, M., Ding, R., Denlinger, R. H., Wong, A., Smeal, T. (2010). Evaluation of selective gamma‐secretase inhibitor PF‐03084014 for its antitumor efficacy and gastrointestinal safety to guide optimal clinical trial design. Molecular Cancer Therapeutics, 9(6), 1618–1628.
Weijzen, S., Rizzo, P., Braid, M., Vaishnav, R., Jonkheer, S. M., Zlobin, A., Miele, L. (2002). Activation of Notch‐1 signaling maintains the neoplastic phenotype in human Ras‐transformed cells. Nature Medi- cine, 8(9), 979–986.
Weng, A. P., Millholland, J. M., Yashiro‐Ohtani, Y., Arcangeli, M. L., Lau, A., Wai, C., Tobias, J. (2006). c‐Myc is an important direct target of Notch1 in T‐cell acute lymphoblastic leukemia/lymphoma. Genes and Development, 20(15), 2096–2109.
Yang, M. H., Wu, M. Z., Chiou, S. H., Chen, P. M., Chang, S. Y., Liu, C. J., Wu, K. J. (2008). Direct regulation of TWIST by HIF‐1alpha promotes metastasis. Nature Cell Biology, 10(3), 295–305.
Yoshida, R., Nagata, M., Nakayama, H., Niimori‐Kita, K., Hassan, W., Tanaka, T., Ito, T. (2013). The pathological significance of Notch1 in oral squamous cell carcinoma. Laboratory Investigation, 93(10), 1068–1081.
Yuan, X., Wu, H., Han, N., Xu, H., Chu, Q., Yu, S., Wu, K. (2014). Notch signaling and EMT in non‐small cell lung cancer: Biological significance and therapeutic application. Journal of Hematology and Oncology, 7, 87. Zanaruddin, S. N. S., Yee, P. S., Hor, S. Y., Kong, Y. H., Ghani, W. M. N. W. A., Mustafa, W. M. W., Cheong, S. C. (2013). Common oncogenic mutations are infrequent in oral squamous cell carcinoma of Asian origin. PLoS One, 8(11), e80229.
Zhao, W., Jie, C., Wei, Z., Yang, Z., Wang, Z., Liu, L., Bing, Q. (2016). Axon guidance molecule semaphorin3A is a novel tumor suppressor in head and neck squamous cell carcinoma. Oncotarget, 7(5), 6048–6062.
Zheng, Y., Wang, Z., Ding, X., Dong, Y., Zhang, W., Zhang, W., Song, X. (2017). Combined Erlotinib and PF‐03084014 treatment contributes to synthetic lethality in head and neck squamous cell carcinoma. Cell Proliferation, 51, 12424.