Abstract
Background: Despite improved treatment for gastric cancer (GC), the prognosis of advanced disease remains poor. Further investigation of the oncogenic sequence for GC is needed. Materials and Methods: The expression of TYRO3 protein tyrosine kinase in five GC cell lines was confirmed using western blotting. TYRO3 knockdown in GC cells, and bromodeoxyuridine and Transwell assays were used to examine the functions of TYRO3 in tumor proliferation and invasion. Finally, TYRO3 expression in 138 patients who underwent curative gastric resection for advanced GC (Union for International Cancer Control stage II/III) was tested by immunohistochemistry, and the association between prognosis and TYRO3 expression was analyzed. Results: TYRO3 was detected at various levels in all the tested GC cell lines. Deleting TYRO3 significantly suppressed proliferation and invasion. Immunohistochemistry revealed TYRO3 expression was an independent prognostic factor for overall survival in patients with GC. Conclusion: TYRO3 appears to mediate tumor progression and predict prognosis of patients with GC.
Gastric cancer (GC) is one of the most common cancers worldwide (1, 2). The highest GC incidence and mortality rates are reported in Eastern and Western Asia, Latin America, and some former Soviet European countries (3). If treated early, the current 5-year overall survival (OS) rate for GC is over 90%: nearly 100% for those with mucosal tumors, and 80-90% for those with submucosal tumors (4-7). However, patients with advanced GC still have the poorest prognosis.
Several different steps can be discerned in oncological cascades, and the engagement of several molecules [nuclear factor-kappa B, extracellular signal-regulated kinase (ERK), phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT), SNAI1 snail family transcriptional repressor 1, and hypoxia-inducible factor] in GC has been suggested (8). Furthermore, immunotherapeutic approaches, such as inhibitors of programmed death receptor-1 and programmed cell death 1-ligand 1, have been added to conventional cytotoxic chemotherapies. However, satisfactory outcomes for patients with advanced GC have not yet been obtained. Therefore, further investigation of the oncogenic sequence for GC is needed.
TYRO3 protein tyrosine kinase, AXL receptor tyrosine kinase and MER proto-oncogene tyrosine kinase (MERTK) comprise the TAM family of receptor tyrosine kinases (RTKs) and were identified in the 1990s (9-11). The family contains common domain structures and are related via a KWIAIES sequence (12). The activation patterns of TAM RTKs are unusual; for maximal stimulation, extracellular lipids and bridging proteins ligand are needed.
According to previous reports, TAM RTKs seem to be ectopically or excessively expressed in different human cancer tissues, and they mediate oncogenicity, including tumor cell survival, chemoresistance, migration, and invasion (13, 14). Moreover, although TYRO3 reportedly plays a role in oncogenicity and may be a therapeutic target in colonic cancer (15) and pancreatic cancer (16), there are only a few reports have suggested that TYRO3 acts as an oncogene in tumors, such as thyroid cancer (17) and melanoma (18). The function of TYRO3 in GC has not been extensively assessed. Assuming the various functions of TAM RTKs in cancer, targeting all TAM RTKs induced in several cancer types may lead to an entirely novel strategy for cancer treatment.
We assess the function of TYRO3 as an oncogene in proliferation of and invasion by GC cell lines and the prognostic significance of TYRO3 overexpression in advanced GC in the present study.
Materials and Methods
Cell culture. The expression of TYRO3 was examined in five human GC cell lines: MKN-1, MKN-45 and MKN-74, which were purchased from the Riken Cell Bank (Tsukuba, Japan); and KATO-III and MKN-28 cells, which were kindly provided by Prof. Ito H (Division of Organ Pathology, Tottori University, Japan). All of the cell lines were cultured in Dulbecco's modified Eagle's medium (Nissui Pharmaceutical, Tokyo, Japan) and supplemented with 10% fetal bovine serum.
Western blotting. The commercially available antibodies used for western blotting were anti-TYRO3 (#5585; Cell Signaling Technology, Danvers, MA, USA) and anti-β-actin (#47778; Santa Cruz Biotechnology, Dallas, TX, USA). For western blotting, the methods used have been described previously (16). The protein signals were detected with Immobilon Western Chemiluminescent Substrate (Millipore, Billerica, MA) and quantified using the Image Quant LAS 4000 mini (GE Healthcare, Chicago, IL, USA).
Cell invasion assay. As described previously, invasion by tumor cells was evaluated using a Transwell assay (8-μm pore size; Corning, Corning, NY, USA) (16).
Gene silencing of TYRO3 using siRNA. Before gene silencing by siRNA transfection, MKN-28 GC cells (1×106 cells) were seeded into 6-well plates using normal medium. According to the manufacturer's protocol, cells were transfected and depleted of TYRO3 expression using siTYRO3 reagent (#36438; Santa Cruz Biotechnology), including three different siRNA duplexes, or with control siRNA reagent (#37007; Santa Cruz Biotechnology), and incubated for 24 h. After replacing the fresh medium, the cells were cultured for 48 h and then analyzed.
Bromodeoxyuridine (BrdU) incorporation assay. BrdU was purchased from Tokyo Chemical Industry Co. Ltd. (Tokyo, Japan). Mouse monoclonal antibody to BrdU was purchased from Medical & Biological Laboratories Co., Ltd. (Nagoya, Japan). We seeded 1×105 cells into a 24-well plate and incubated it at 37°C for 24 hours. BrdU incorporation was evaluated after a 2-h pulse, as described previously (16).
Patients and immunohistochemical staining. In this analysis, we included 138 patients with advanced GC (stage II/III by eighth edition of the Union for International Cancer Control classification) (19) who underwent curative gastric resection at Tottori University Hospital (Yonago, Japan) between February 2003 and December 2011. We excluded patients who underwent re-resection for residual cancer. The Tottori University Ethical Board approved the protocol (approval number: 20A054). Clinicopathological parameters and laboratory data of all patients extracted from their electronic medical records were used in the study. In detail, patient characteristics (age, sex, body mass index, cancer type) and surgical and perioperative parameters [neoadjuvant chemotherapy, adjuvant chemotherapy, pathological stage, histological type, vascular invasion, pathological T-stage, lympho-node metastasis, carcinoembryonic antigen (CEA), and carbohydrate antigen 19-9] were analyzed retrospectively.
The sections (4-μm-thick) from paraffin-embedded tissues were stained using anti-TYRO3 (#PAB3391; Abnova, Neihu District, Taipei City) and counterstained with hematoxylin in order to depict nuclei. We considered that cells with strong staining of the cytoplasm or plasma membrane were positive for TYRO3 protein expression; strong staining was defined as staining intensity equal to that of the islet of Langerhans cells within non-cancerous pancreatic tissues, which were stained at the same time, as external positive controls (16). For analysis, we used a two-tiered discrimination using positive or negative staining. Tumors with >10% positive cells were classified as TYRO3-positive. Two investigators (U. C. and M.M.) evaluated the immunolabelling; agreement was obtained in each case.
Statistical analysis. Univariate analyses were performed using Fisher's exact test for categorical variables and two-tailed t-tests for continuous variables. The Kaplan–Meier method was used for survival curves, and the differences in survival curves were compared using the log-rank test. Cox proportional hazards models were used for the multivariate analysis. All statistical analyses were examined using SPSS v. 23.0 statistical software (IBM Corporation, Armonk, NY, USA), and p-values less than 0.05 were considered statistically significant.
Results
Expression of TYRO3 in GC cell lines. To confirm the expression of TYRO3 in GC, we first evaluated TYRO3 expression using five human GC cell lines (KATO-III, MKN-1, MKN-28, MKN-45 and MKN-74). In all the examined cell lines, TYRO3 expression was detected at various levels (Figure 1A).
Effect of silencing of TYRO3 in GC cells in vitro. As TYRO3 is potentially an oncogene (15, 17, 18), we performed TYRO3 knockdown with specific siRNA to confirm whether TYRO3 is involved in tumor progression in GC. We then examined the effect of TYRO3 knockdown in MKN-28 cells. Cell proliferation was assessed by incorporation of BrdU in the cells after a 2-hour pulse. Compared with the control, depletion of TYRO3 significantly inhibited BrdU incorporation (Figure 1B). Additionally, a Transwell assay was performed to evaluate the effects of TYRO3 depletion on GC cell invasion. The results demonstrated that the depletion of TYRO3 significantly reduced the invasive ability of MKN-28 GC cells (Figure 1C). These results indicate that TYRO3 affects GC progression, including cell proliferation and invasion.
TYRO3 expression is associated with poor prognosis in GC. Based on our in vitro studies above, we retrospectively analyzed records and specimens of 138 patients who underwent gastric resection for GC, to identify the prognostic significance of TYRO3 expression in these patients. Immunohistochemical analysis of TYRO3 expression in their GC specimens revealed that TYRO3 expression was positive in 58% (80/138) of the samples (Figure 2A, Table I). TYRO3 expression was significantly stronger in well-differentiated adenocarcinoma than moderately/poorly differentiated adenocarcinoma (Table I) but the other factors had no significant relationship to its expression.
TYRO3 protein tyrosine kinase expression in gastric cancer (GC) cell lines, and effects of TYRO3 deletion on GC cell proliferation and invasion. A: TYRO3 expression in GC cell lines was analyzed using western blotting. B: Cell proliferation of MKN-28 cells was evaluated through the bromodeoxyuridine incorporation assay. C: Cell invasion by MKN-28 cells was analyzed with a Transwell assay. Scale bar: 200 μm. ctrl: Control; DAPI: 4’,6-diamidino-2-phenylindole. *Statistically significant at p<0.05.
TYRO3 protein tyrosine kinase expression in gastric cancer (GC) tissues is negatively correlated with prognosis of patients with GC. A: Representative images of GC tissue samples immunohistochemically stained for TYRO3. Lower: Image of cells in an islet of Langerhans in noncancerous pancreatic tissue used as a positive control. Scale bar: 100 μm. B: Survival curves of patients with GC with negative (n=58) and positive (n=80) TYRO3 expression. *Statistically significant at p<0.05.
Association between TYRO3 protein tyrosine kinase expression and clinicopathological factors.
We then evaluated whether TYRO3 expression had an important role as a prognostic risk factor in patients with GC. In univariate analysis, age ≥65 years, body mass index ≥25 kg/m2, administration of neoadjuvant chemotherapy, stage III disease, pN-positive, serum CEA level ≥5 ng/ml, and positive TYRO3 expression were risk factors for poorer overall survival (OS) after surgery in this cohort (Table II). The Kaplan–Meier survival analysis revealed that patients with TYRO3-positive GC had significantly worse OS (p=0.005), disease-specific survival (p=0.001), and recurrence-free survival (p=0.002) compared with TYRO3-negative patients (Figure 2B). In multivariate analysis by Cox proportional hazards model, age, neoadjuvant chemotherapy, stage and TYRO3 expression were independent prognostic factors for OS of patients with GC (Table III). Based on the results of these studies, TYRO3 overexpression is a significant negative prognostic factor for patients with GC.
Prognostic significance of clinicopathological characteristics.
Discussion
Multidisciplinary approaches to diagnosing and treating patients with GC have been widely evaluated (20-22) but current strategies have not attained satisfaction for those with advanced GC. In order to improve prognosis of these patients, novel therapeutic strategies are needed.
Some oncogene and tumor-suppressor mutations have been found in GC. For example, activating Kirsten rat sarcoma viral oncogene homolog mutations are found in approximately 5-20% of GCs (23). Based on molecular mechanisms underlying GC, researchers have investigated various targeted therapies belonging to different classes of drugs as therapeutics in GC, starting with preclinical studies, such as human epidermal growth factor receptor type 2, vascular endothelial growth factor receptor, the insulin-like growth factor receptor, PI3K/AKT pathway, c-mesenchymal–epithelial transition factor, fibroblast growth factor receptor and immunotherapies (24).
The TAM RTKs are not always excessively expressed in individual cancer types (12), this indicates that each TAM family member may have different functions in different tumor types. BGB324, a selective inhibitor of AXL currently in phase II clinical trials for cancer, is a hopeful therapy for pancreatic cancer (25). In addition, the MERTK inhibitor named UNC2025 had antitumor effects in preclinical models of melanoma (26). However, data on selective inhibitors of TYRO3 in any type of tumor have been limited. Moreover, although TYRO3 is reportedly oncogenic and may be a therapeutic target for colon cancer (15) and pancreatic cancer (16), few studies have intimated that TYRO3 has any oncogenic potential in solid tumors (17, 18). Considering their different interactions and roles, targeting TAM RTKs induced in each cancer type might bring about a novel anticancer strategy.
Unlike AXL and MERTK, the function of TYRO3 in tumors, including GC, has not been widely examined. We therefore investigated the role of TYRO3 in GC progression. The expression of TYRO3 protein was detected in all the examined human GC cell lines (KATO-III, MKN-1, MKN-28, MKN-45 and MKN-74). Interestingly, in our cell-based assays, TYRO3 knockdown inhibited GC cell proliferation as well as invasion.
Multivariate analysis of prognostic factors for overall survival in patients with gastric cancer.
The TAM family has two major ligands, growth arrest-specific factor 6 (GAS6) (27, 28) and protein S (PROS1) (29), which bind to the receptor with their carboxyl-terminal domains. GAS6 and PROS1 also bind to phosphatidylserine (PS) with their amino terminus, thus connecting TAM receptors between PS. PROS1 is known to be a true ligand for MERTK and TYRO3, but not for AXL (12). AXL also binds to GAS6 with the strongest affinity and can be sufficiently activated through receptor dimerization by GAS6, even without PS (29). Cancer cells have also been shown typically to present increased PS levels on their plasma membrane surfaces (30). Being overexpressed, TYRO3 can create a homodimer or heterodimer and become auto-phosphorylated at the tyrosine residue in the kinase domain, even without ligands (31, 32). Together, these findings suggest that TYRO3 is activated in a ligand-specific fashion or through RTK dimerization, even in GC.
Epithelial–mesenchymal transition (EMT), in which tightly connected epithelial cells dissociate and become motile and invasive mesenchymal cells, is a critical stage in GC progression (8). In GC, several signaling pathways regulate EMT, most notably PI3K/AKT, mitogen-activated protein kinase kinase/ERK and WNT/β-catenin (8). Interestingly, reported downstream mediators of TAM are common to these EMT regulators (12, 16), which implies that TYRO3 may promote GC progression via some signaling cascades, including EMT.
Our immunohistochemical analyses of advanced GC specimens obtained from 138 patients showed that tumors from 80 (58%) patients were positive for TYRO3 expression. We confirmed TYRO3 overexpression to be significantly correlated with poorer OS in patients with GC, as an independent prognostic factor. As far as we know, this is the first report to confirm an inverse correlation between the expression of TYRO3 and the prognosis of patients with stage II/III GC.
As a conclusion, this study shows that TYRO3 is constitutively induced in GC cells and can mediate the progression of tumor. We also found, for the first time, that the expression of TYRO3 predicts prognosis of patients with stage II/III GC. Determining TYRO3 expression in GC may help predict patient prognosis, and TYRO3 or its downstream molecules may provide a novel therapeutic target against GC.
Acknowledgements
This study was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number 18K16312 (MM) and 20K17689 (WM). The Authors also thank the Edanz Group (https://en-author-services.edanzgroup.com/ac) for editing a draft of this article.
Footnotes
Authors' Contributions
Chihiro Uejima: Conceptualization, software, data curation, writing - original draft, validation. Masaki Morimoto: Conceptualization, methodology, software, data curation, writing - review and editing. Manabu Yamamoto: Conceptualization, data curation. Kazushi Hara: Data curation. Wataru Miyauchi: Resources. Ken Sugezawa: Data curation. Yoichiro Tada: Data curation. Akimitsu Tanio: Resources, data curation. Kyoichi Kihara: Data curation. Tomoyuki Matsunaga: Data curation. Naruo Tokuyasu: Data curation. Teruhisa Sakamoto: Data curation. Soichiro Honjo: Data curation. Yoshihisa Umekita: Supervision, methodology. Yoshiyuki Fujiwara: Supervision. Conceptualization, project administration, writing - review and editing.
Conflicts of Interest
The Authors have no conflicts of interest or financial ties to disclose.
- Received July 17, 2020.
- Revision received August 8, 2020.
- Accepted August 11, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.