Autocrine Leukemia Inhibitory Factor Promotes Esophageal Squamous Cell Carcinoma Progression via Src Family Kinase-Dependent Yes-Associated Protein Activation

This article is featured in Highlights of This Issue, p. 1757

Esophageal cancer is the eighth most common cancer worldwide and the sixth most common cause of death from cancer (1). Although there have been recent advances in multimodal therapies such as chemotherapy, radiotherapy, immunotherapy, and surgery, the prognosis of patients with esophageal cancer is still poor because of its aggressive clinicopathologic features (2). In Japan, esophageal squamous cell carcinoma (ESCC) accounts for approximately 90% of all histologic types of esophageal cancer (3). ESCC frequently exhibits lymph node metastasis and tumor invasion into adjacent organs, even in the early stages, contributing to poor prognosis (4).

Inflammation is a complex biological response to cellular damage resulting from cell injury or infection, in which the immune system attempts to eliminate damaging stimuli and initiate healing and regenerative process (5–7). Inflammatory responses also play pivotal roles in cancer development, including tumor initiation, promotion, progression, and metastasis (7, 8). IL6 family members, including IL6, IL11, IL27, IL31, leukemia inhibitory factor (LIF), and oncostatin M, are one of the best characterized protumorigenic cytokines that affect cell proliferation, survival, inflammation, and metabolism (9). Most IL6 family members activate the JAK–STAT3, Src homology 2 (SHP-2)–Ras–Raf–MEK–ERK, and PI3K–AKT pathways, as well as the SFK–Yes-associated protein (YAP) pathway via the common signaling receptor subunit, glycoprotein 130 (gp130; refs. 10–12). The IL6–STAT3 signaling pathway has been shown to be activated in ESCC, evoking cascades that lead to epithelial–mesenchymal transition (EMT; ref. 13). YAP is frequently activated in human ESCC; however, its mechanisms of activation remain unknown, although some ESCCs show YAP amplification (14, 15).

LIF is a pleiotropic cytokine of the IL6 family that plays important roles in the maintenance of self-renewal and totipotency of embryonic stem cells (16, 17). The effects of LIF differ according to cell type and include stimulation or inhibition of cell proliferation (16). In cancer cells, LIF has been shown to stimulate growth, inhibit differentiation, and induce metastasis in a variety of tumors (18). In human esophageal adenocarcinoma (EAC), LIF expression is reportedly associated with treatment resistance (19). LIF expression is observed in human EAC tissues, and silencing of LIF reduces cell viability in human EAC cell lines (20). However, the pathophysiologic role of LIF in human ESCC progression remains unclear.

The current study investigated the pathophysiologic role of LIF in ESCC development using both human ESCC cell lines and human ESCC clinical samples. In human ESCC cell lines, LIF knockdown reduced cell proliferation, migration/invasion and sphere formation, and induced apoptosis. Similar to IL6, LIF was shown to regulate the SFK–YAP pathway via modulation in the stability and intracellular localization of YAP, and subsequent transcription of YAP target genes, as well as the JAK–STAT3 pathway. LIF was also found to be expressed in human ESCC resected samples. Concomitant inhibition of the SFK–YAP and JAK–STAT3 pathways may represent an innovative therapy for human ESCC.

Among the 136 patients who underwent esophageal resection at the Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University (Fukuoka, Japan) between 2008 and 2012, 111 patients with ESCC who underwent curative resection and whose cancer tissue was pathologically confirmed were included in the current study. Written informed consent was obtained from all patients or their guardians and all specimens were collected with the approval of the ethics committee and the Institutional Review Board at the Kyushu University Hospital (approval number: 29-444). The research conforms to the provisions of the Declaration of Helsinki. ESCC tissue microarray, containing 96 cases of human ESCC, was obtained from US Biomax, Inc.

Female SCID Beige mice were purchased from Charles River Laboratories Japan Inc. and were kept in specific pathogen-free facilities at Keio University (Tokyo, Japan). TE-11 cells (1 × 10⁷ cells) with LIF or control knockdown by shRNAs in 200 μL of PBS–Matrigel (1:1) mix were subcutaneously injected using a 26-G needle into a 6-week-old female SCID Beige mouse. Tumor sizes were monitored every 2 days with a digital caliper. The tumor-bearing mice were euthanized at 5 weeks after the injection, and the tumor tissues were collected, fixed in 4% paraformaldehyde (PFA; Nacalai Tesque), and used for histologic examination. Hematoxylin and eosin (H&E) staining was performed using Tissue-Tek hematoxylin 3G and eosin (Sakura Finetek Japan) according to the manufacturer's instructions. The images were taken using BZ-X810 microscope (KEYENCE). Animal experiments were performed in strict accordance with the Guidelines for Proper Conduct of Animal Experiments of the Science Council of Japan. All experiments using mice were approved by the Animal Ethics Committee of Keio University (Approval number: 08004) and were performed according to the Animal Ethics Committee guidelines.

Recombinant human IL6 and LIF were obtained from PeproTech Inc., and Matrigel (Product Number 354234) for tumor implantation was obtained from Corning Inc. Mitomycin C was obtained from Kyowa Kirin. AZD0530 and AZD1480, which are SFK and JAK inhibitors, respectively, were obtained from Selleck Chemicals. YAP inhibitors, CIL56 (21) and verteporfin (22), were obtained from Cayman Chemicals.

IHC was performed using 5-μm-thick paraffin-embedded tissue sections. The sections were deparaffinized in xylene and dehydrated in a series of ethanol concentrations ranging from 100% to 70%. For antigen retrieval, specimens in citrate buffer (pH 6.0) or Target Retrieval Solution, pH 9.0 (DAKO) were heated in an autoclave at 110°C or microwave at 99°C for 15 minutes. The sections were incubated for 30 minutes in 3% hydrogen peroxide to deactivate endogenous peroxidases. After blocking the nonspecific binding of antibodies with Blocking One Histo (Nacalai Tesque), the specimens were incubated with primary antibodies at 4°C overnight. Antibody labeling was achieved using a DAKO EnVision Detection System (Dako) or horseradish peroxidase (HRP)–conjugated secondary antibodies. The sections were reacted with 3,3′-diaminobenzidine, counterstained with hematoxylin, and dipped in 0.2% ammonia water for bluing, and mounted. The images were taken using NanoZoomer (Hamamatsu Photonics KK) and BZ-X700 or BZ-X810 microscope (KEYENCE). Primary antibodies included antibodies against YAP (Cell Signaling Technology, #14074, 1:200), phospho-Src family (Tyr416; Cell Signaling Technology, #2101, 1:100), LIF (R&D Systems, #AF-250-NA, 1:100), Ki67 (Cell Signaling Technology, #9449, 1:200), and cleaved caspase-3 (Cell Signaling Technology, #9661, 1:200).

The human ESCC cell lines, TE-1, TE-5, TE-8, and TE-11, were obtained from the Cell Resource Center for Biomedical Research, Tohoku University (Sendai, Japan), and were cultured in RPMI1640 (Nacalai Tesque) supplemented with 10% FBS, and 1% penicillin/streptomycin, 10 mmol/L HEPES, and 1 mmol/L sodium pyruvate. Human embryonic kidney 293T cells were obtained from ATCC and cultured in DMEM supplemented with 10% FBS, 1% penicillin/streptomycin, 10 mmol/L HEPES, and 1 mmol/L sodium pyruvate. Mycoplasma testing was performed using MycoAlert Mycoplasma Detection Kit (Lonza) or BioMycoX Mycoplasma PCR Detection Kit (CellSafe), and all the cell lines used in this study were found to be Mycoplasma negative. All cell lines were authenticated by short tandem repeat profiling, which was performed by Takara Bio Inc. Cells were used within five cell passages and the assays were conducted using cells within two months of thawing.

LIF-specific siRNAs were synthesized by Sigma-Aldrich. The siRNAs specific for LIF were MISSION siRNAs named SASI_Hs01_00130401 and SASI_Hs01_00130402. MISSION siRNA Universal Negative Control #1 (siNC; SIC-001, Sigma-Aldrich) was used as a nontargeting siRNA. Cells were seeded in 6-well plates (2.0 × 10⁵ cells/well) and reverse transfected with 50 nmol of LIF siRNAs or siNC using Lipofectamine RNAiMAX Transfection Reagent (Thermo Fisher Scientific) for 72 hours according to the manufacturer's instructions.

The lentiviral packaging plasmids psPAX2 (Addgene plasmid #12260) and pMD2.G (Addgene plasmid #12259), which were gifts from Dr. Didier Trono (Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland), were cotransfected for lentiviral production with human LIF-specific (TRCN0000058587, TRCN0000058585, and TRCN0000058586, Sigma-Aldrich), human LIFR-specific (TRCN0000058772 and TRCN0000058770, Sigma-Aldrich), or human YAP-specific (23) pLKO1 shRNA plasmids, FLAG-tagged YAP5SA (an active form of YAP; ref. 24)–expressing plasmid, or FLAG-tagged STAT3C (an active form of STAT3; ref. 25)–expressing plasmid. Luciferase shRNA (shLuc), targeting firefly luciferase, was used as a nontargeting shRNA. Transfection into HEK293T cells was performed using polyethylenimine MAX (PEI MAX, Polysciences, Inc.), and after 24 hours of the transfection, the medium was changed to remove the transfection reagents. The virus-containing medium was harvested 48 and 72 hours after the initial transfection. For transduction into cells, virus-containing medium was filtered (0.45-μm pore size) and then infected with 8 μg/mL polybrane (Nacalai Tesque). The viral infected cells were selected using puromycin (InvivoGen) or G418 (Nacalai Tesque) before use.

Equal amounts of total, nuclear, or cytoplasmic protein from each sample were separated using SDS-PAGE and blotted onto polyvinylidene difluoride membranes. NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific) were used for the extraction of separate nuclear and cytoplasmic protein fractions. Protein blots were hybridized with the indicated primary antibody of interest and followed by HRP-conjugated secondary antibody, followed by detection with Chemi-Lumi One L or Chemi-Lumi One Super (Nacalai Tesque) and the FUSION Chemiluminescence Imaging System (VILBER). Primary antibodies included antibodies against YAP (Cell Signaling Technology, #14074, 1:1,000), phospho-YAP (Ser127; Cell Signaling Technology, #4911, 1:1,000), phospho-Src family (Tyr416; Cell Signaling Technology, #2101, 1:1,000), Src (Cell Signaling Technology, #2110, 1:1,000), phospho-STAT3 (Tyr705; Cell Signaling Technology, #9145, 1:1,000), STAT3 (Cell Signaling Technology, #9132, 1:1,000), CTGF (Cell Signaling Technology, #10095, 1:1,000), CYR61 (Cell Signaling Technology, #39382, 1:1,000), phospho-YAP (Tyr357; Abcam, #ab62751, 1:3,000), α-tubulin (Sigma-Aldrich, #T9029, 1:1,000), Lamin B1 (Invitrogen, #33-2000, 1:1,000), and FLAG (DYKDDDDK; Wako Chemicals, #019-22394, 1:1,000).

Adherent cells plated on Lab-Tek Chamber slide (Thermo Fisher Scientific) were washed with PBS and fixed in 4% PFA for 15 min. The slides were blocked for 1 hour with 5% goat serum and 1% BSA in PBS. Primary antibodies were added to the cells and incubated at 4°C overnight. Alexa Fluor 546 goat anti-rabbit antibody (Thermo Fisher Scientific, 1:300) was used as the secondary antibody. Coverslips were then mounted on the slides using VECTASHIELD Mounting Medium with DAPI (Vector Laboratories). Primary antibodies included antibodies against YAP (Cell Signaling Technology, #14074, 1:100) and phospho-STAT3 (Tyr705; Cell Signaling Technology, #9145, 1:100).

Total RNA was extracted using RNAiso Plus (Takara Bio Inc.) and reverse transcribed using a High-Capacity cDNA Synthesis Kit (Thermo Fisher Scientific). Real-time PCR analysis was performed using the CFX Connect Real-Time PCR Detection System (Bio-Rad) and the SsoFast EvaGreen Supermix (Bio-Rad). All primer sets yielded a single product of the expected size. Relative expression levels were normalized to 18S rRNA. The sequences of the primers used in this study are shown in Supplementary Table S1.

To examine IL6 and LIF secretion into the media, human IL6 uncoated ELISA Kit with plates (Thermo Fisher Scientific) and human LIF ELISA Kit (Thermo Fischer Scientific) were used. The experiments were performed according to the manufacturer's instructions.

Cell proliferation ability was measured by seeding cells into 96-well plates at an initial density of 2,000 cells per well. Every 24 hours after seeding for 4 to 6 days, 10 μL of the Cell Count Reagent SF (Nacalai Tesque) was added to the cell culture medium for 1 or 2 hours, and the absorbance for each well was measured at a 450-nm wavelength using an iMark microplate reader (Bio-Rad).

Apoptosis was analyzed using an Annexin V-FITC Apoptosis Detection Kit (BioVision) following double staining with FITC-conjugated Annexin V and propidium iodide (PI). Annexin V binds to apoptotic cells with exposed phosphatidylserine, whereas PI binds to the late apoptotic and necrotic cells with membrane damages. Staining was performed according to the manufacturer's instructions. Stained cells were analyzed on a FACS Canto II (BD Biosciences), and the data was analyzed using Diva (BD Biosciences) and FlowJo (Tree Star) softwares.

Cells were incubated in 6-well plates until they reached 100% confluence. A scratch was then made in the monolayer with a micropipette tip to create a “wound,” and then, cells were treated with Mitomycin C (3 μg/mL) for 1 hour. Micrographs of the wound areas were taken at 0, 24, and 48 hours after scratching using a BZ-X700 microscope (KEYENCE). The percentage wound closure was quantified and analyzed using ImageJ software.

Falcon permeable supports for 24-well plate with 8.0-μm Transparent PET Membrane were placed in the 24-well plate and 5,000 cells per insert were seeded in basal medium. The bottom chamber was filled with basal medium supplemented with 10% FBS as a chemoattractant. Cells were allowed to migrate for 24 hours and then fixed and stained using Diff-Quik (Sysmex). Micrographs of the migrated cells were taken using a BZ-X700 microscope (KEYENCE) and migrated cells per 10× field were counted and analyzed.

Corning BioCoat Matrigel Invasion Chambers with 8.0-μm PET membranes were placed in the upper chamber with 5,000 cells per insert seeded in basal medium. The bottom chamber was filled with basal media supplemented with 10% FBS as a chemoattractant. Cells were allowed to migrate for 24 hours and then fixed and stained using Diff-Quik (Sysmex). Micrographs of the migrated cells were taken using a BZ-X700 microscope (KEYENCE) and invaded cells per 10× field were counted and analyzed.

To evaluate the primary spherical formation, TE-11 cells from subconfluent monolayer cultures were plated at a density of 5 × 10⁴ in a 24-well untreated plate (Thermo Fisher Scientific) using 3D Tumorsphere Medium XF (PromoCell). Micrographs of the tumor spheres were taken using BZ-X700 or BZ-X810 microscope (KEYENCE) after 5 to 7 days and the number of tumor spheres per 10× field was counted and analyzed.

All statistical analyses, data processing, and graph generation were performed using GraphPad Prism 6.0 or 8.0. All values are presented as mean ± SEM. Student t test was used to compare two groups, whereas ANOVA multivariate analysis was performed for all other comparisons. Statistical significance was determined as P < 0.05. Overall survival (OS) was defined as the time between the day of surgery to the day of death or the most recent follow-up visit. Kaplan–Meier curves were used to estimate the distributions of OS, log-rank test was used to determine statistical significance, and χ² test and Pearson correlation method were used to compare the IHC results.

We first investigated whether STAT3, SFK, and YAP were activated in TE-1, TE-5, TE-8, and TE-11 human ESCC cell lines by measuring expression of phosphorylated protein and total protein, as well as expression of YAP target genes, such as CCN2/CTGF, CCN1/CYR61, amphiregulin (AREG), and JAG1, and confirmed the activation of these molecules (Fig. 1A and B). We next measured the expression of IL6 family cytokines in the human ESCC cell lines using real-time PCR. IL6 and LIF were expressed at various levels in all four cancer cell lines, whereas IL11 and IL22 were barely detected (Fig. 1C). With consistent the real-time PCR results, both IL6 and LIF proteins were detected by ELISA in the culture supernatants of human ESCC cell lines (Fig. 1D). The common receptor for IL6 family cytokines, IL6ST/gp130, as well as IL6 receptor (IL6R) and LIF receptor (LIFR), were also expressed in human ESCC cell lines (Fig. 1E), suggesting that IL6 and LIF may establish a functional autocrine loop in human ESCC. We confirmed that human ESCC cell lines were able to respond to exogenous IL6 and LIF by monitoring STAT3 activation (phosphorylation and nuclear translocation of STAT3) after IL6 and LIF stimulation (Fig. 1F; Supplementary Fig. S1A). These data suggest that IL6 and LIF may function in an autocrine manner in human ESCC cell lines.

To investigate the role of autocrine LIF in human ESCC cell lines, LIF was knocked down in the ESCC cells using two different siRNAs or three different shRNAs against LIF, and knockdown efficiency was confirmed by real-time PCR (Fig. 2A; Supplementary Fig. S2A and S2B; Supplementary Fig. S3B). LIF knockdown suppressed the expression of YAP1/YAP, WWTR1/TAZ, and their target genes (Fig. 2B; Supplementary Fig. S2B and S2C) and also suppressed nuclear translocation of both YAP and STAT3 (Fig. 2C and D; Supplementary Fig. S2D). Exogenous LIF stimulation induced SFK and STAT3 activation, YAP tyrosine phosphorylation (Y357), accumulation, and nuclear translocation although exogenous LIF stimulation did not affect YAP serine phosphorylation (S127), which is phosphorylated by the Hippo signaling pathway components Lats1 and Lats2 (Fig. 2E and F; ref. 26). LIF-induced YAP nuclear translocation was inhibited by the addition of AZD0530 (1 μmol/L), an SFK inhibitor (Fig. 2F). These findings suggest that LIF activates YAP via SFK activation in human ESCC cells probably through a Hippo pathway-independent mechanism.

Next, we investigated the role of LIF in human ESCC cell proliferation and apoptosis. LIF knockdown suppressed ESCC cell proliferation and expression of proliferation-related genes, such as CCND1/Cyclin D1, MYC/c-Myc, and MKI67 (Fig. 3A and B; Supplementary Fig. S3A and S3B). Consistent with the findings from the LIF knockdown experiment, knockdown of LIFR also suppressed ESCC cell proliferation (Supplementary Fig. S3C). Moreover, LIF knockdown induced ESCC cell apoptosis and reduced the expression of antiapoptosis genes, such as BCL2/Bcl-2 and BCL2L1/Bcl-_XL (Fig. 3B and C; Supplementary Fig. S3D and S3E). These data suggest that LIF critically contributes to human ESCC cell proliferation and apoptosis.

We also investigated the role of LIF in cell migration and invasion. LIF knockdown prevented cell migration in a wound healing assay, suppressed cell migration and invasion in the Transwell assays, and inhibited expression of EMT-related transcriptional factors, such as SNAI2/SLUG, SNAI1/SNAIL, and TWIST1/TWIST (Fig. 4A,–D; Supplementary Fig. S4A). These findings suggest that LIF plays essential roles in regulating human ESCC cell migration and invasion in addition to cell proliferation and apoptosis. Interestingly, LIF expression was induced under a sphere-forming condition, and LIF knockdown also inhibited sphere formation in the human ESCC cells and suppressed the expression of SOX2, a cancer stem cell marker expressed in ESCC (Fig. 4E,–G; ref. 27). These data are suggestive of the important role of LIF in cancer stem cells.

We investigated whether YAP or STAT3 is more important downstream of LIF using siRNAs against LIF and overexpression of active forms of YAP and STAT3 (Fig. 5A and B). We found that LIF-induced YAP activation is more important for cancer cell proliferation than LIF-induced STAT3 activation in human ESCC cells (Fig. 5C,–E). Furthermore, constitutively active YAP and STAT3 coexpressed in LIF-knockdown TE-8 cells could fully rescue cell growth (Fig. 5E). These data suggest that the LIF–SFK–YAP pathway plays a crucial role in human ESCC growth and the SFK–YAP pathway cooperatively functions with the JAK–STAT3 pathway for cell proliferation during the downstream of LIF signaling.

Next, we determined whether these findings could be reproduced in human clinical ESCC samples. We investigated the relationship between YAP activation (nuclear YAP) and SFK activation (phosphorylated SFK), as well as YAP activation and LIF expression in esophageal cancer cells by immunostaining human ESCC clinical samples (Fig. 6A,–E). SFK activation was associated with YAP activation in human ESCC samples and patients with ESCC with YAP activation showed a poor prognosis (Fig. 6A,–C). The expression level of LIF was positively but weakly correlated with the activation level of YAP in a human tissue microarray analysis of patients with ESCC (Fig. 6D and E), suggesting that together with LIF signaling, other signaling pathway(s), such as IL6 signaling, might also play a role in YAP activation in vivo. Taken together, these data indicate the importance of the LIF–SFK–YAP pathway in human ESCC clinical samples.

We investigated the role of YAP in human ESCC cells using either shRNAs against YAP or YAP inhibitors. Consistent with the previous report (14), YAP knockdown severely inhibited the growth and expression of proliferation-related genes of human ESCC cell lines (Supplementary Fig. S5A and S5B), as was also observed in the LIF-knockdown experiments (Fig. 3A and B; Supplementary Fig. S3A and S3B). The two YAP inhibitors used (CIL56 and verteporfin) also pharmacologically inhibited the YAP pathway (21, 22). We found that the YAP inhibitors exhibited an anticancer effect in the human ESCC cell lines, reducing cell viability in a dose-dependent manner and sphere formation as well as the expression of YAP target genes (Fig. 7A; Supplementary Fig. S5C and S5D). These data further suggest the importance of YAP as a potential target for human ESCC treatment.

Next, we investigated the therapeutic effect of the concomitant inhibition of the SFK–YAP and JAK–STAT3 pathways by SFK and JAK inhibitors, some of which are clinically being used, as these pathways are independent and activated by LIF in human ESCC cell lines. As expected, concomitant inhibition of these pathways effectively suppressed cell proliferation and expression of proliferation-related genes in human ESCC cell lines compared with single pathway inhibition (Fig. 7B; Supplementary Fig. S6A). Furthermore, 5-fluorouracil (5-FU), a standard therapy for human ESCC, in combination with these inhibitors, additively suppressed cell proliferation (Fig. 7C). These findings suggest that targeting both the SFK–YAP and JAK–STAT3 pathways may represent a novel therapeutic strategy for human ESCC.

Finally, we investigated the role of autocrine LIF signaling during in vivo tumor growth because murine LIF is incapable of activating human LIFR (28). We subcutaneously transplanted TE-11 cells with LIF and control knockdown by shRNAs into SCID Beige mice. The LIF-knockdown ESCC cells grew slower than the control cells in the SCID Beige mice (Fig. 7D). Consistent with the result, the tumor histologic analysis showed less proliferative (Ki67-positive) tumor cells and more apoptotic (cleaved caspase-3-positive) tumor cells as well as less YAP activation in LIF-knockdown tumors (Fig. 7E; Supplementary Fig. S6B). These data suggest that autocrine LIF signaling serves as a major inducer of ESCC growth in vivo.

We investigated the role of autocrine LIF in human ESCC cells, as it is highly expressed in these cells. LIF knockdown suppressed proliferation, migration/invasion, and sphere formation and induced apoptosis of cancer cells. LIF activated YAP via SFK activation as well as the JAK–STAT3 pathway, and LIF-induced YAP activation was shown to be more important for cancer cell proliferation than LIF-induced STAT3 activation. Furthermore, we revealed that SFK activation and LIF expression were correlated with YAP activation in human ESCC tissue and autocrine LIF signaling enhanced human ESCC growth in vivo. Therefore, the LIF–SFK–YAP axis may represent a novel target for human ESCC therapy.

LIF is a member of the IL6 family of cytokines and plays a pivotal role in the maintenance of self-renewal and pluripotency of embryonic stem cells (16). LIF expression was also induced under sphere-forming condition in our study, suggesting the importance of LIF in the maintenance of cancer stem cells. In addition, LIF promotes tumor development in autocrine or paracrine manners in various cancers (16, 29). LIF functions in an autocrine manner and several types of cancer cells reportedly express LIF after acquiring mutations (30–32). However, the transcription factors, signaling pathways, and/or epigenetic changes involved in the induction of LIF expression in cancer cells remain unknown. It is well known that LIF can activate the JAK–STAT3 pathway via gp130. In the current study, we showed that, similar to IL6, LIF can also activate the SFK–YAP pathway in cancer cells, and concomitant inhibition of the SFK–YAP and JAK–STAT3 pathways effectively suppressed ESCC growth. Because JAK inhibitors and SFK inhibitors are currently used to treat other diseases, early clinical application is expected. Moreover, the addition of JAK and SFK inhibitors to 5-FU additively suppressed the proliferation of ESCC cells because the activation of STAT3 and YAP is closely associated with chemoresistance (9, 33).

We also found that the SFK–YAP pathway was more important in human ESCC cell proliferation than the JAK–STAT3 pathway. JAK inhibitors are used clinically to suppress STAT3 activity; however, their long-term use may lead to immunodeficiency, which may increase the risk of malignancy and infection (34). Therefore, in some cases, it may be advised to only target the SFK–YAP pathway and avoid the JAK–STAT3 pathway.

Studies into the relationship between LIF and IL6 have reported that these cytokines may be redundant in promoting ovarian cancer (35). Our results indicate that IL6 is unable to compensate for the role of LIF in ESCC cells, because LIF knockdown induced a severe phenotype despite IL6 still being expressed. LIF and IL6 activate similar downstream signaling pathways, such as the JAK–STAT3, Ras–ERK, PI3K–Akt, and SFK–YAP pathways (10–12). Therefore, further studies are required to investigate why IL6 is unable to compensate for LIF in human ESCC cells.

In conclusion, our findings revealed the expression of LIF and its importance in human ESCC cells. LIF promotes tumor development by activating STAT3 and YAP, which are suppressed by JAK inhibitors and SFK inhibitors, respectively. Systemic STAT3 inhibition by JAK inhibitors results in immunosuppression; therefore, elucidating the downstream signaling pathways of LIF is an important consideration for LIF-targeted cancer treatment.

E. Oki reports honoraria for lecturing from Taiho Pharmaceutical Co., Ltd., Yakult Honsha Co., Ltd., Merck Biopharm., Takeda Pharmaceutical Co., Ltd., Chugai Pharmaceutical Co., Ltd., Bayer, Eli Lilly, and Ono Pharmaceutical Co., Ltd. No potential conflicts of interest were disclosed by the other authors.

T. Kawazoe: Formal analysis, investigation, writing-original draft, writing-review and editing. H. Saeki: Resources, writing-review and editing. E. Oki: Resources, writing-review and editing. Y. Oda: Resources, writing-review and editing. Y. Meahara: Resources, writing-review and editing. M. Mori: Resources, writing-review and editing. K. Taniguchi: Conceptualization, resources, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing-original draft, project administration, writing-review and editing.

We thank Drs. A. Yoshimura, S. Chikuma, and D. Aki for helpful discussion, and Y. Noguchi, Y. Hirata, Y. Tokifuji, Y. Honma, R. Ogawa, H. Watanabe, C. Miyamoto, T. Sato, M. Yoshikawa, A. Nohmi, and K. Yoshinaga for technical assistance. We also thank Y. Tanishita for manuscript preparation. This work is supported by JSPS KAKENHI (JP 15K21775, JP 20H03758), AMED (PRIME) under grant number JP 18gm6210008/19gm6210008/20gm6210008, the “Kibou Projects” Startup Support for Young Researchers in Immunology, the Keio Gijuku Academic Development Funds, the Uehara Memorial Foundation, the Kanae Foundation for the Promotion of Medical Science, the Astellas Foundation for Research on Metabolic Disorders, the SENSHIN Medical Research Foundation, research grants from Bristol-Myers Squibb, the SGH foundation, the MSD Life Science Foundation, the Ichiro Kanehara Foundation for the Promotion of Medical Sciences and Medical Care, the Yasuda Medical Foundation, the Suzuken Memorial Foundation, the Pancreas Research Foundation of Japan, the Waksman Foundation of Japan Inc., the Japanese Foundation for Multidisciplinary Treatment of Cancer, Project Mirai Cancer Research Grants from the Japan Cancer Society, the Okinaka Memorial Institute for Medical Research, the Asahi Glass Foundation, the Foundation for Promotion of Cancer Research, the Kobayashi Foundation for Cancer Research, the Toray Science Foundation, the Vehicle Racing Commemorative Foundation, the JSR-Keio University Medical and Chemical Innovation Center (JKiC), a research grant from the Public Trust Surgery Research Fund, a Research Grant of the Princess Takamatsu Cancer Research Fund, the Tokyo Biomedical Foundation, the Daiichi Sankyo Foundation of Life Science, the Mochida Memorial Foundation for Medical and Pharmaceutical Research, the Medical Research Encouragement Prize of the Japan Medical Association, the Terumo Foundation for Life Sciences and Arts, the Yakult-Bioscience Foundation, the Novartis Foundation, the Mitsubishi Foundation, and the Takeda Science Foundation (all to K. Taniguchi), and JSPS KAKENHI (JP 19J11357; to T. Kawazoe). This work was also supported by JSPS KAKENHI grant number JP 16H06279 (PAGS) and JSPS KAKENHI grant number JP 16H06276 (AdAMS).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.