KRT17 Facilitates Proliferation, Migration and Invasion in Esophageal Squamous Cell Carcinoma by Regulating mTOR/S6K1 pathway

Esophageal squamous cell carcinoma (ESCC) is one of the most aggressive malignancies worldwide which originates from the malignant transformation of esophageal epithelial cells. Dysregulated expression of Keratin17 (KRT17) has been claimed in a variety of malignancies, while its role in ESCC remains unclear. Therefore, our study aimed to explore the potential function and underlying molecular mechanism of KRT17 in ESCC. Methods Data-independent acquisition-mass spectrometry (DIA-MS) workow was used to analysis KRT17 expression between ESCC and adjacent non-cancerous esophageal tissues. The online database gene expression proling interactive analysis (GEPIA) was used to further determine the differential expression of KRT17 in tissues. The function of KRT17 in ESCC was tested on two human esophageal cancer cell lines (EC9706 and ECA109). Small interfering RNA (siRNA) was used to inhibit KRT17 expression. Cell proliferation was examined by cell counting kit 8 (CCK8) reagent, colony formation assay, cell cycle distribution analysis and apoptosis. Cell migration was examined by transwell and wound healing assay. Cell invasion was also examined by transwell assay. Western blot and quantitative real-time PCR (qRT-PCR) was used to evaluate protein and mRNA levels, respectively.


Background
Esophageal cancer is a leading cause of cancer-related deaths in the world with a low 5-year survival rate, about 19% [1]. Esophageal cancer can histologically classify into two predominate subtypes, esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC) [2], while ESCC is more prevalent than EAC, account for approximately 90% of cases with esophageal cancer worldwide [3]. Despite rapid progress in medicine, many ESCC patients are still diagnosed at an advanced stage.
Patients who receive prompt treatment also suffer from local or systemic recurrence and have a poor prognosis [4]. Hence, it is meaningful to investigate the pathogenesis and underlying molecular mechanism of ESCC, which may aim of developing better strategies to improve ESCC prevention, management and treatment.
Keratin17 (KRT17) belongs to type intermediate lament, which molecular weight is 48 KDa [5]. KRT17 is involved in multiple biological processes, represented by skin in ammation and cell proliferation [6]. The dysregulated expression of KRT17 has reported in several types of cancer, including gastric, bladder, pancreatic and colorectal cancer [7]. However, the role of KRT17 in the ESSC has not been elucidated.
The mammalian target of rapamycin (mTOR) is an atypical serine/threonine kinase which plays vital roles in cell growth, metabolism, autophagy and immunity [8]. mTOR is frequently dysregulated in the pathogenesis of cancer, thus it has the potential to be a therapeutic target [9]. S6 kinase 1 (S6K1) is encoded by RPS6KB1 on chromosome 17 and can phosphorylated and activated by mTOR [10,11]. The overexpression of S6K1 is closely related with worse prognosis and metastasis in multifarious human cancers, represented by breast, colorectal, lung and ovarian cancer [12]. The mTOR-S6K1 pathway has been considered to be an irreplaceable regulator in many pathological processes.
In the present study, we aimed to determine the expression of KRT17 in tissues and explore the potential function and underlying molecular mechanism of KRT17 in the pathogenesis of ESCC. We showed that KRT17 was overexpression in ESCC and knockdown of KRT17 attenuated the proliferation, migration and invasion of ESCC cell lines by inhibiting mTOR/S6K1 pathway.

Materials And Methods
Patients and Specimens 27 pairs of human ESCC tumor tissues and adjacent non-cancerous esophageal tissues were collected from patients with ESCC at Sichuan Provincial People's Hospital from May to November in 2019. None of the patients received radiotherapy, chemotherapy, or other anticancer treatment before surgery. The specimens were collected immediately following radical resection of esophageal carcinoma and ash frozen in liquid nitrogen. The staging of esophageal cancer should be performed according to the tumornode-metastasis (TNM) classi cation published by the current American Joint Council on Cancer (AJCC).
The histological features of all specimens were evaluated by pathologists according to the WHO criteria.
Informed consent was collected from each participant. All the methods in this study were in accordance with approved guidelines, and this study was approved by the Medical Ethics Committee of Sichuan Provincial People's Hospital.

Tissue sample preparation
We extracted total protein with a standard protocol. Frozen tissue was scraped and added with SDT lysate to transfer to Lysing Matrix A tube. The BCA protein assay (Beyotime, China) was used for protein quanti cation. The samples were stored at -80℃. For each sample, 20μg of protein was added into 6X sample loading buffer, and the samples were bathed in boiling water for 5min. 12% SDS-PAGE electrophoresis (constant pressure 250V, 40min) was performed, and the samples were stained with Coomase blue. Protein samples were digested according to the manufacturer's protocol for lter-aided sample preparation (FASP) [13].
Agilent 1260 In nity HPLC System were used to fractionate the peptides of all samples. The sample was loaded onto the chromatographic column (Waters, XBridge Peptide BEH C18 Column, 130Å, 5 μm, 4.6 mm X 100 mm) equilibrated with the buffer A (5% ACN, 10 mM, pH 10.0) and separated at the ow rate of 1 mL/min. Liquid chromatography separation was performed using linear gradient from the buffer B (85% ACN, 10mM HCOONH4, pH 10.0) concentration of 5% to 45% over 40 minutes at a constant column temperature of 30℃. Each fraction was dried in a vacuum concentrator. After lyophilization, 0.1% formic acid aqueous solution was used to re-dissolved the samples.
Via Spectronaut Pulsar X (version 12, Biognosys AG), mass spectrometry RAW data were combined, analyzed, searched and established a spectrogram database named Uniprot_HomoSapiens_20386_20180905 with download link http://www.uniprot.org in June, 2018. Trypsin was set as the enzyme allowing two missed cleavage sites. The search parameters were as follows: xed modi cation of Carbamidomethyl (C) and variable modi cations of Oxidation(M) and acetyl (Protein N-term). The standards were 1% Precursor FDR, 1% Protein FDR and 1% Peptide FDR.
DIA mass spectrometry analysis 6ul of each sample was added to 1μL 10×iRT peptide segment. After mixing, 6μL of each sample was injected and separated by Nano-LC. Then the sample was analyzed by online electrospray tandem mass spectrometry. The whole liquid-mass tandem system was as follows: 1) Liquid phase system: Easy nLC system (Thermo Fisher Scienti c); 2) mass spectrometry system: Q-Exactive HF-X (Thermo Fisher Scienti c). The sample was separated by nonlinear growth gradient at a ow rate of 300nL/min on analytical columns (Thermo Fisher Scienti c, Acclaim PepMap RSLC 50um X 15cm, Nano Viper, P/N164943). Total RNA was extracted from cells by using TRIzol reagent (Invitrogen, USA) according to the manufacturer's protocol. Reverse transcription reaction was conducted with 37℃ for 15min, subsequent 85℃ for 5s. qRT-PCR reaction were performed by using the SYBR® PrimeScript™ RT-PCR Kit (Takara, Japan) and GAPDH served as the internal control. The fold changes were calculated by using the 2 −ΔΔCt method. Primer sequences were listed as follow, KRT17 forward 5 -GGTGGGTGGTGAGATCAATGT-3 and reverse 5 -CGCGGTTCAGTTCCTCTGTC-3 ; GAPDH forward 5 -GGACCTGACCTGCCGTCTAG-3 and reverse 5 -GTAGCCCAGGATGCCCTTGA-3 .

Western blot
Total protein was extracted from cells by using a mixture composed of radioimmunoprecipitation lysis buffer, phenylmethylsulfonyl uoride (Beyotime, China) and phosphatase inhibitor (Solarbio, China), then separated by SDS-PAGE and transferred onto a PVDF membrane (Millipore, USA). Transferred membranes were blocked with 5% non-fat milk for 1h at room temperature and then incubated overnight at 4℃ with following primary antibodies, KRT17, mTOR, p-mTOR, mTOR, S6K1, p-S6K1 (1:1000, Proteintech, USA) and β-actin (Bioss, China), followed by incubated with the HRP labeled goat antibody against rabbit IgG (Beijing Zhongshan Biotechnology Co. Ltd., China) for 1h at room temperature. After that, the density of protein bands was captured by chemiluminescence and analyzed by ImageJ (NIH, USA) Cell counting kit 8 (CCK8) The 96-well plates were precultured in the incubator for 24 hours with 100 μl cell suspension per well, then added 10μl substance with different concentrations. Incubated plates in the incubator for an appropriate period of time, add 10μl CCK8 solution (Solarbio, China) to each well and put plates in the incubator for 1-4 hours. Plates were measured by a microplate reader (Biotek, USA) at a corresponding wavelength.

Colony formation assay
Seeded with 1000 cells/well in 6-well plates and continued to culture for 14 days, then xed with 4% paraformaldehyde and stained with 0.1% crystal violet. Cell number in each well were counted by an optical microscope (Nikon, Japan).

Wound healing assay
Transfected cells were seeded into 12-well plates until cell con uence >90%, a 10μl sterile pipette tip was scratched along the longitudinal direction of the cavity, replaced plates with new medium (0.5% FBS). Images of cell migration were captured at 0h, 24h and 48h by an inverted microscope.

Transwell migration and invasion assay
Both migration and invasion capacity can be assessed by using transwell inserts. Transfected cells were seeded into the upper chambers of inserts and were cultured at 37℃. Then inserts were inverted on absorbent paper to wipe the medium and remove no-migratory cells with a cotton swab. After that, cells were xed with 4% paraformaldehyde and stained with 0.1% crystal violet, counted in 3 random elds per sample.

Statistical analysis
All statistical analyses were performed with SPSS 22.0 software (SPSS Inc., Chicago, IL, USA). Data were presented as mean ± SD. One-way ANOVA or Student's t-test was used for the analyses. P<0.05 were considered statistically signi cant.

KRT17 was overexpression in ESCC
We collected human ESCC tumor tissues and adjacent non-cancerous esophageal tissues to screen out the differentially-expressed proteins (DEPs). In Fig. 1A, signi cantly up-regulated and down-regulated proteins were shown in the volcano plot, and the number of DEPs were listed in Fig. 1B, the volcano plot of DEPS of ESCC tumor (T) and non-cancerous esophageal tissues (N) groups is based on differences in protein expression. The heatmap of the 1200 signi cantly differentially expressed proteins was shown in Fig. 1C, blue represents the downregulation and red indicates upregulation. The Gene Ontology contains biological process, molecular function and cellular components, which can detailedly and accurately demonstrate gene products function. The top 20 of biological process, molecular function and cellular components were shown in Fig. 1D, Fig. 1E and Fig. 1F, respectively. A total of 7344 terms were enriched for the DEPs, and the amount of DEP enriched in each term were listed in Fig. 1G. The above results revealed the crucial DEP, KRT17.
To further demonstrate the role of KRT17 in ESCC, we explored KRT17 expression according to online database. Our study showed that compared to normal tissues, KRT17 was overexpression in tumor tissues ( Fig. 2A), and the overall survival was also closely related with KRT17 levels (Fig. 2B). The results implied us that KRT17 may act as a tumor promoter, its inhibition may improve the status of ESCC.

siKRT17 inhibited the overexpression of KRT17 in ESCC cell lines
To explore the effect of siKRT17 in KRT17 expression, its mRNA and protein levels were measured in EC9706 and ECA109. Fig. 2C showed that both EC9706 and ECA109 KRT17 mRNA level in siKRT17 group was signi cantly lower than that in NC group. Western blot results were corresponded with qRT-PCR assay, as shown in Fig. 2D, siKRT17 down-regulated KRT17 expression in both cell lines. These results suggested that siKRT17 effectively down-regulated KRT17 expression in EC9706 and ECA109.

KRT17 inhibition attenuated proliferation, migration and invasion in ESCC cell lines
To determine the role of KRT17 in the pathogenesis of ESCC, we evaluated KRT17 function on proliferation, migration and invasion in EC9706 and ECA109. The effect of KRT17 on cell proliferation was veri ed by cell viability, CCK8 assay showed that compared with siCtrl group, cell lines transfected with siKRT17 had slower cell proliferation (Fig. 3A). Cell proliferation was also indicated by colony formation assay, as shown in Fig. 3B, the number of cell clones in siKRT17 group was reduced obviously (P<0.01), and the analysis of EC9706 and ECA109 are consistent. The change of DNA content can re ect cell cycle, compared with siCtrl group, EC9706 and ECA109 siKRT17 group had lower cell count in S and G2/M phase, while had higher cell count in G1 phase (Fig. 3C). Moreover, we measured the percentage of apoptotic cells to re ect cell proliferation laterally, EC9706 and ECA109 transfected with siKRT17 had a higher apoptosis rate. These results strongly suggested that KRT17 facilitated ESCC cell proliferation.
Tumor cells can acquire migration and invasion phenotype to accelerate cancer progression, and KRT17 was overexpression in ESCC based on previous results. Hence, we explored KRT17 function in cell migration and invasion. As shown in Fig. 4A, compared with siCtrl group, wound healing assay showed that the migration rate in siKRT17 group was reduced at 48h in EC9706 and ECA109. Transwell assay revealed that KRT17 inhibition in both cell lines effectively reduced the invasion capacity (Fig. 4B). We further used transwell assay to assess cell migration, the result showed that EC9706 and ECA109 transfected with siKRT17 had a lessened migration capacity (Fig. 4C).

KRT17 regulated proliferation, migration and invasion via mTOR/S6K1 pathway
The mTOR pathway was reported to play a role in proliferation, migration and invasion [14]. The activation of mTOR brought about S6K1 phosphorylation, and mTOR/S6K1 pathway is also associated with angiogenesis [15]. To investigate the role of mTOR/S6K1 pathway in KRT17 related proliferation, migration and invasion, we used western blot to detect the protein levels of mTOR and S6K1. The results showed that EC9706 and ECA109 transfected with siKRT17 signi cantly decreased mTOR (Fig. 5A) and S6K1 (Fig. 5B) phosphorylation levels.

Discussion
ESCC is the main subtype of esophageal cancer with high mortality in the world [16]. Despite the remarkable advances in diagnostic methods and treatments including surgical excision, chemotherapy and radiotherapy in recent years, the situation of ESCC patients remains unsatis ed with most cases diagnosed at an advanced stage and existing treatments have not brought about long-term survival bene ts [17]. The relatively low survival rate has led to extensive research on ESCC aimed at changing the poor status, while the molecular mechanisms in the pathogenesis of ESCC still not clear. KRT17 belongs to type intermediate lament with multiple functions, deregulated expression of KRT17 has been shown to be related with the pathogenesis of diversi ed cancers [18]. Recent study showed that KRT17 was overexpression in osteosarcoma, and KRT17 inhibition reduced osteosarcoma cell proliferation via regulating the AKT/mTOR/HIF1α pathway [19]. KRT17 was also up-regulation in cervical cancer, knockdown of KRT17 attenuated proliferation and migration, and stimulate cervical cancer cell apoptosis [20]. In gastric cancer, KRT17 function as a promoter to facilitate tumor growth, motility and invasion [21]. However, the role of KRT17 in ESCC has not been clari ed.
The present study aimed to identi ed the potential function and underlying molecular mechanism of KRT17 in ESCC. First of all, we needed to determine KRT17 expression. DIA-MS is a proteomic methodology for deep and proteome-wide pro ling with high accuracy and reproducibility in proteomic quanti cation [22,23], we used DIA-MS to investigate the DEPs, and the result showed that the expression of KRT17 was up-regulated in ESCC tumor tissues, the differential expression was further determined by the data obtained from GEPIA. Also, the overall survival of ESSC patients was also closely related with KRT17 levels. These results implied that KRT17 potentially act as an oncogene in the pathogenesis of ESCC, knockdown of KRT17 can better explore its function, EC9706 and ECA109 transfected with siKRT17 inhibit the expression of KRT17 by the evidence of signi cantly decreased protein and mRNA levels.
CCK8 assay can directly re ect the role of KRT17 on cell proliferation, the result showed that knockdown of KRT17 induced by siRNA manifested a lower proliferation capacity in EC9706 and ECA109. Consistent with CCK8 assay, siKRT17 group had a fewer number of cell clones. In EC9706 and ECA109 transfected with siKRT17, the percentage of cells in S and G2/M phase decreased while the percentage of cells in G1 phase increased when compared to that transfected with siCtrl. Also, EC9706 and ECA109 transfected with siKRT17 enhance the proportion of apoptotic cells. Transwell and wound healing assay still revealed negative effects of siKRT17 on migration and invasion. All in all, knockdown of KRT17 in EC9706 and ECA109 led to the alleviation of proliferation, migration and invasion capacity, these results adequately demonstrated the close relationship between KRT17 and these cell viability functions. mTOR is associated with diverse molecular and biological aspects of cancer, and overexpression of mTOR can promote tumor growth [24]. As a mTOR downstream effector, S6K1 overexpression is associated with poor cancer prognosis [25]. Growing studies have focused on the relationship between mTOR/S6K1 pathway and cancer. Our present study revealed that KRT17 inhibition decreased the phosphorylation levels of mTOR and S6K1 in EC9706 and ECA109. These results showed that knockdown of KRT17 attenuated proliferation, migration and invasion of ESCC cell lines by inhibiting mTOR/S6K1 pathway.

Conclusion
Our study demonstrated that KRT17 was overexpression in ESCC tissues. Knockdown of KRT17 led to the alleviation of proliferation, migration and invasion capacity in ESCC cell lines, and these processes was mediated by inhibiting mTOR/S6K1 pathway. KRT17 may be a novel and effective target for ESSC treatment. Figure 1 (A) Volcano plot of differentially expressed proteins between T and N groups. The Fold change and Pvalue of protein expression between the two groups of samples were used, the x-coordinate is the difference multiple (logarithmic transformation based on 2) and the y-coordinate is the P-value  (C) qRT-PCR analysis of KRT17 mRNA level among NC group and siKRT17 group in EC9706 and ECA109.

Declarations
(D) Western blot analysis of KRT17 protein level among control group, NC group and siKRT17 group in EC9706 and ECA109. Data were shown as mean ± SD (n=3). **P<0.01 versus NC group. Abbreviations: ESCA, esophageal carcinoma; GEPIA, gene expression pro ling interactive analysis; KRT17, keratin17; NC, negative control; qRT-PCR, quantitative reverse transcription polymerase chain reaction. Figure 3 (A) Cell proliferation in EC9706 and ECA109 transfected with siCtrl or siKRT17 were measured by CCK8. (B) Cloning counts in EC9706 and ECA109 transfected with siCtrl or siKRT17. Data were shown as mean ± SD (n=3). **P<0.01 versus siCtrl group. (C) Cell cycle distribution in EC9706 and ECA109 transfected with siCtrl or siKRT17 were detected by FCM. (D) Apoptosis in EC9706 and ECA109 transfected with siCtrl or siKRT17 were detected by FCM. Abbreviations: CCK8, cell counting kit 8; FCM, ow cytometry. Figure 4 (A) Cell migration rate at 0h, 24h and 48h in EC9706 and ECA109 transfected with siCtrl or siKRT17 were measured by wound healing assay. (B) Cell invasion in EC9706 and ECA109 transfected with siCtrl or siKRT17 were measured by transwell assay. (C) Cell migration rate in EC9706 and ECA109 transfected with siCtrl or siKRT17 were measured by transwell assay. Figure 5 (A) Western blot analysis of p-mTOR and p-S6K1 protein levels in EC9706 and ECA109 transfected with none, siCtrl or siKRT17.