Background
Biliary tract cancer (BTC) is a relatively rare but aggressive gastrointestinal cancer with a high mortality rate [
1]. Surgical resection is the only curative modality for BTC [
2]; although complete resection has been achieved, the recurrence of BTC is high in patients with resected BTC [
3] with a poor 5-year survival of 20–30% [
4,
5]. Furthermore, most tumors metastasize at diagnosis, imposing difficulties in standard treatments [
6]. Therefore, palliative treatments other than systemic chemotherapy or radiotherapy are the only option for patients with unresectable or metastatic BTC. The combined use of gemcitabine and cisplatin is recommended as first-line treatment for these patients [
7]. This combination chemotherapy improves progression-free and overall survival, but the median overall survival is no longer than one year in metastatic BTC [
8].
Cancer stem cells (CSCs) play crucial roles in tumor initiation, progression, and relapse. Most CSCs show characteristics of epithelial-mesenchymal transition (EMT); therefore, CSCs in BTCs are highly desmoplastic, morphologically heterogenic, aggressive, and resistant to chemotherapy. Previous studies have demonstrated that CSCs in BTC are responsible for the low response to anti-cancer treatment and high recurrence rate [
9]. Identifying specific CSC markers for BTC is a dynamic field of research. Several potential markers have been proposed, including CD24, CD44, EpCAM, CD133, and aldehyde dehydrogenase 1 [
10‐
13]. However, the phenotypic heterogeneity and cellular plasticity of CSCs hinder their application [
14]. Therefore, identifying a representative CSC marker for BTC and understanding the characteristics of CSCs (especially EMT) in BTC are crucial to improving chemosensitivity and developing targeted therapy for inhibiting EMT.
Several studies have explored the potential of sphere formation, a well-established method for maintaining cells with stem cell-like properties [
15‐
18]. In our previous study, we performed cDNA microarray analysis using adherent and sphere cells from the human BTC cell lines SNU1196 and SNU245 to evaluate the unique molecular patterns of BTC CSCs and identified 70 genes (1.2 > FC (fold change) in spheres and 2 > FC in BTC) [
19]. Among the identified genes,
Transgelin-2 (TAGLN2) was overexpressed in BTC and CSCs. TAGLN2, an actin-binding protein highly expressed by tumor cells, plays a crucial role in determining cell morphology and transformation [
20] and has been implicated in various human malignancies [
21‐
32]. The potential involvement of TAGLN2 in metastasis, either through direct interaction with cytoplasmic actin or induced expression of metastasis-related genes, has been reported in several studies [
30,
33,
34]. A recent study exploring the role of
TAGLN2 in cancer demonstrated its association with multidrug resistance and metastasis in breast cancer, emphasizing the potential of
TAGLN2-targeted therapies [
35]. Furthermore,
TAGLN2 has been identified as an oncogene related to prognosis and immunity across various cancers [
31], and its involvement in tumor proliferation and migration in colorectal cancer has also been documented [
28].
In this study, we investigated the malignant behavior of TAGLN2 in BTC cells and analyzed its expression in patient tissue and serum. Suppressing TAGLN2 expression by short hairpin RNAs (shRNA) decreased cell motility, tumorigenic ability, down-regulated CSC-related markers, and enhanced cell sensitivity to radiation or chemotherapeutics. In patient samples, TAGLN2 expression was enhanced in both cancer tissues and patient serums. Our data suggest that TAGLN2 is a putative diagnostic marker and therapeutic target for BTC that targets EMT and CSCs.
Materials and methods
Cell culture
The human BTC cell lines SNU245, SNU308, SNU478, SNU869, SNU1079, and SNU1196 were purchased from the Korea Cell Line Bank (KCLB, Seoul, Korea). These cell lines were maintained in RPMI1640 (Invitrogen Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, US). NIH-3T3 mouse fibroblast cells were purchased from the American Type Culture Collection (ATCC Manassas, VA, USA) and maintained in DMEM (Invitrogen, Carlsbad, CA, USA) supplemented with 10% FBS. We previously established gemcitabine-resistant BTC cells by escalating doses of gemcitabine treatment in SNU-1196 cells, as described in a previous study [
19]. Cells were maintained at 37 °C in a humidified incubator with 5% CO
2.
Reagents
Inhibitors and chemotherapy drugs including 5-FU, cisplatin, oxaliplatin, carboplatin, irinotecan, etoposide, erlotinib, crizotinib, and savolitinib were purchased from Selleck Chemicals (Houston, TX, USA); gemcitabine was supplied by Eli Lilly Korea (Seoul, Korea).
Sphere formation assay was performed as described in a previous study [
35]. Briefly, cells were trypsinized and resuspended at a density of 1 × 10
3/well in D/F12 (Invitrogen Gibco) medium supplemented with 10 ng/mL epidermal growth factor (R&D Systems Inc., Minneapolis, MN, USA), 10 ng/mL basic fibroblast growth factor (R&D Systems Inc.), 1X insulin-transferring selenium (Invitrogen), 0.5% bovine serum albumin (Invitrogen), and 0.5% FBS in ultra-low attachment culture plates (Corning Inc., Corning, NY, USA). Adherent cultured cells, as a control, were seeded in culture dishes (Nalgene Nunc Intl, Rochester, NY, USA) with a sphere formation medium. After seven days, cells were collected and dissociated with Accutase (Sigma-Aldrich, St. Louis, MO, USA).
Microarray analysis
Total RNA was extracted from adherent cells and cell spheres of SNU245 and SNU1196 cells using RNeasy Miniprep kits (Qiagen, Valencia, CA, USA). Microarray analysis was performed according to the kit’s protocol [
19]. Briefly, double-stranded cDNA was prepared using 6 µg aliquots of total RNA, amplified using polymerase chain reaction (PCR), and labeled with biotin using an IVT labeling kit (Affymetrix, Santa Clara, CA, USA). The labeled cDNA was fragmented and hybridized to an Affymetrix GeneChip Human Genome U133 Plus 2.0 high-density oligonucleotide Array (Affymetrix). The microarrays were then washed using a GeneChip Fluidics Station 450 (Affymetrix) and scanned using a GeneChip Array Scanner 3000 7G (Affymetrix). Expression data were generated using Affymetrix Expression Console software version 1.1 using MAS5 algorithm normalization. The expression intensity data in the CEL file were normalized using the MAS5 algorithm to reduce noise.
Silencing TAGLN2 expression using shRNA, small interfering RNA (siRNA)
To verify the effects of TAGLN2 inhibition, a TAGLN2 knockdown cell line was established by transfecting TAGLN2 shRNA (targeting sequence: 5′-GTGCTATGTGAGCTCATTAAT-3′) or mock shRNA (targeting sequence: 5′-GGAATCTCATTCGATGCATAC-3′) plasmids (Sure Silencing shRNA plasmids; cat. 336,314 KH19252P) into SNU1196 cells using Lipofectamine 2000 (Invitrogen), followed by treatment with puromycin (2 µg/mL). Single colonies were picked, and TAGLN2 expression was assessed using western blotting. TAGLN2 siRNA (Santa Cruz Biotechnology, Santa Cruz, CA; sc-106,633) comprising a pool of a target-specific siRNA (siTAGLN2) and control siRNA-A (Santa Cruz Biotechnology; sc-37,007) at 100 pmol/L in 150 nM medium were transfected into SNU1196/GR cells using Lipofectamine RNAiMAX (Invitrogen).
Semiquantitative reverse-transcription PCR (RT-PCR)
Total RNA was extracted for RT-PCR using RNAeasy Mini kits (Qiagen), and single-stranded cDNA was synthesized using a Superscript II system (Invitrogen), according to the manufacturer’s instructions. The expression of TAGLN2 was evaluated in gastric cancer spheres and control cells, and the Actin beta (ACTB) gene was used as the reference gene. The following primers were used for RT-PCR: TAGLN2 forward primer: 5′-TAT GGC ATT AAC ACC ACT GA-3 ′; TAGLN2 reverse primer: 5′-GGA TTC TCC TTG GAT TTC TT-3 ′; ACTB forward primer: 5′-GGC ATC CTC ACC CTG AAG TA-3′; ACTB reverse primer: 5′-GGG GTG TTG AAG GTC TCA AA-3′.
Western blotting
Western blotting was performed as described in a previous study [
19]. Briefly, cells were homogenized in lysis buffer (70 mM glycerophosphate, pH 7.2, 0.6 mM Na vanadate, 2 mM MgCl
2, 1 mM EGTA, 1 mM DTT, 0.5% Triton X-100, 0.2 mM phenylmethylsulfonyl fluoride, and 1X complete protease inhibitor; Roche Applied Science, Nutley, NJ, USA), incubated on ice for 1 h, centrifuged at 12,000 rpm for 1 h at 4 °C, and the supernatant was collected. Recombinant TAGLN2 (rTAGLN2) was generated by AbFrontier (Seoul, Republic of Korea) by expressing the full-length
TAGLN2 sequence in the pET21a vector in BL21 cells for three days, and the cells were collected. Cells were washed with phosphate-buffered solution (PBS) and cultured with serum-free RPMI (Invitrogen Gibco) for two days for protein precipitation from culture media. The culture supernatant was collected, mixed with acetone, and stored at − 20℃ overnight. The mixture was centrifuged at 13,000 rpm for 30 min at 4 °C, and the pellet was collected and resuspended in lysis buffer. Patient blood samples were collected in 10 mL BD serum tubes, centrifuged at 4 °C for 20 min at 3,000 × g, and 3 µL of the supernatant serum samples were loaded. Primary antibodies against TAGLN2 (Sigma-Aldrich), N-cadherin (Abcam, Cambridge, UK), Snail, Nanog, JAG2, phospho-GSK3, phospho-AKT, phospho-MEK, MAPK, phospho-ERK (Cell Signaling Technology, Inc., Danvers, MA, USA), cMET, occludin, AKT, MEK, ERK, and GAPDH (Santa Cruz Biotechnology) were used. The secondary antibodies used were goat anti-mouse HRP and goat anti-rabbit immunoglobulin G (IgG)-HRP (Santa Cruz Biotechnology; 1:5000). Proteins were visualized using Super Signal® West Pico Chemiluminescent Substrate (Thermo Scientific, Rockford, IL, USA).
Cell proliferation assay
The cell proliferation assay was performed using WST-1 reagent (Roche Applied Science) according to the manufacturer’s instructions. SNU-1196 cells transfected with TAGLN2 or control shRNA were seeded into 96-well plates at 1 × 103/well in 100 µL culture medium. After 48 h, WST-1 (10 µL) was added to each well and incubated at 37 °C for 1 h. Absorbance was measured at 450 nm using a VersaMax ELISA microplate reader (Molecular Devices, USA).
Migration and invasion assays
Migration and invasion assays were performed as described in a previous study [
35]. For the migration assay, cells were detached and suspended at 1 × 10
5 cells/mL in serum-free media and plated at a density of 1 × 10
4 cells/well in 24-well Transwell plates (Costar, Bethesda, MD). For the invasion assay, the upper chamber was pre-coated with Matrigel (1:4 diluted with serum-free medium; BD Biosciences), and cells were seeded at a density of 1 × 10
4 cells/well. The bottom chamber was filled with the culture medium containing NIH-3T3 fibroblasts. The cells were incubated for 24 and 72 h for migration and invasion assays, respectively. After incubation, cells were fixed with 5% glutaraldehyde for 30 min and stained with 0.1% crystal violet. The cells were completely removed from the upper surface of the membrane using a moist cotton swab. The migrated and invaded cells were counted and photographed under a microscope at 100× magnification. All assays were performed in triplicates.
Colony formation assay was performed as described in a previous study [
35]. Briefly, 3 × 10
2 cells were suspended in 0.5 mL Difco Noble Agar (0.3%; Becton Dickinson Company, Sparks, MD) supplemented with sphere formation medium and plated in 24-well plates containing 0.6% agar. All samples were plated in triplicates. Cells were incubated for 2–3 weeks in a humidified incubator with 5% CO
2 at 37 °C, and the sphere formation medium was changed every alternative day. Radioresistance was analyzed by subjecting cells to 0–8 Gy of ionizing radiation (Gammacell 3000 Elan, MDS Nordion, Ottawa, ON, Canada), followed by a colony formation assay.
Cell viability assay
Cells were seeded at in 96-well plates and treated with inhibitors and chemotherapy drugs at various concentrations for 72 h. Cell viability was evaluated using the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) assay (AMRESCO, Solon, OH, USA) [
19]. The half-maximal inhibitory concentration (IC
50) was analyzed relative to that of the DMSO control. Values are shown as the mean of triplicate wells from three independent experiments for each drug concentration. Absorbance was measured at 570 nm using a VersaMax ELISA microplate reader (Molecular Devices). Dose-effect data for individual drugs and their combinations were analysed for synergism using CompuSyn software (
http://www.combosyn.com/).
Tumorigenicity assays
Tumorigenicity assay was performed as described in a previous study [
19]. Briefly, cells were washed with PBS, suspended in serum-free RPMI (Invitrogen Gibco) and Matrigel (BD Biosciences PharMingen; 1:1 volume), and subcutaneously injected into the right flank of six-week-old female BABL/c nude mice (Orient Bio, Seongnam, Korea) [
19]. The tumor volume was calculated as V (mm
3) = (A
2×B)/2, where A is the diameter perpendicular to the largest dimension, B. After 14–16 weeks, the mice were sacrificed in a CO
2 chamber, and tumor tissues were fixed in 4% paraformaldehyde. For histological evaluation, tissue samples were embedded in paraffin and stained with hematoxylin and eosin (H&E). All animal experiments were approved by the Committee for the Care and Use of Laboratory Animals of Yonsei University College of Medicine, and the study is reported in accordance with ARRIVE guidelines(
https://arriveguidelines.org).
Patients
We analyzed 41 human BTC tissue samples obtained from surgical resections and 139 human blood samples from 89 patients with BTC, 10 patients with biliary stones, and 40 normal controls at Severance Hospital, Yonsei University College of Medicine. The Ethical Committee for Clinical Research of the Institutional Review Board of Severance Hospital, Yonsei University College of Medicine, Seoul, Korea, approved the study protocol (IRB approval code:4-2011-0625; November 24, 2011). All procedures involving human participants were performed in accordance with the ethical standards of the Institutional Research Committee and the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Written informed consent was obtained from all subjects. Information regarding patient demographics and clinical data were obtained from electronic medical records, including age at diagnosis, sex, tumor stage at diagnosis, and serum carcinoembryonic antigen (CEA) levels. Tumors were staged according to the 7th edition of the American Joint Committee on Cancer (AJCC) staging classification.
Immunohistochemical and immunofluorescence staining
Immunohistochemical and immunofluorescence staining were performed as described in a previous study [
35]. For immunohistochemical staining, tissue slides were deparaffinized in xylene and rehydrated in graded alcohol. Endogenous peroxidase activity was blocked with 0.3% (v/v) hydrogen peroxide in methanol. Antigen retrieval was performed by microwaving the slides in a sodium citrate buffer (0.01 M, pH 6.0) for 5 min. To block nonspecific staining, sections were incubated with 10% (v/v) normal donkey serum for 1 h; then, the sections were incubated with appropriate antibodies overnight at 4 °C. Subsequent reactions were performed using Envision kits (Dako Cytomation California, Inc., Carpinteria, CA, USA) following the manufacturer’s instructions. Immunoreactions were developed with the DAKO Liquid diaminobenzidine substrate-chromogen system (DAB+) and counterstained with Harris hematoxylin (Sigma-Aldrich). The reaction was subsequently carried out with an LSAB + Kit (Dako), and sections were counterstained with Mayer’s hematoxylin, dehydrated, and observed under a BX51 microscope (Olympus, Tokyo, Japan). For immunofluorescence staining, tissue slides were visualized using Cy5-goat anti-rabbit IgG dissolved in an antibody diluent and incubated for 30 min at room temperature [
35]. Between each step, three washing steps of 5 min each were performed on a rocking platform using PBS. The slides were cover-slipped using a mounting medium for observing fluorescence with DAPI (Vecta shield H-1200; Vector Laboratories, Inc. Burlingame, CA, USA). Primary antibodies against TAGLN2 (Sigma-Aldrich), alpha-smooth muscle actin (α-SMA), fibroblast-associated protein (FAP), and cytokeratin-7 (CK-7) (Santa Cruz Biotechnology) were used.
Statistical analysis
Categorical data were analyzed using χ2 and Fisher’s exact tests. Student’s t-test and Mann–Whitney U test were used for continuous variables. Survival was estimated and compared using Kaplan–Meier analysis with a log-rank test. Serum TAGLN2 and CA19–9 levels were compared between patients with benign and biliary cancer using the non-parametric Kruskal–Wallis test. The cut-off value, receiver operating characteristic (ROC) curve, area under the ROC curve (AUC), and 95% confidence intervals (CI) were determined. All statistical analyses were performed using IBM SPSS Statistics for Windows (version 25.0; IBM Corp., Armonk, NY). A P-value of < 0.05 was considered statistically significant.
Discussion
CSCs are a subset of cells within a tumor that possess the unique ability to self-renew, resist chemotherapy, and initiate tumor growth. The discovery of CSCs has led to a new understanding of tumorigenesis and the development of new treatments. Targeting CSCs, in addition to conventional chemotherapy targeting cancer cells, is crucial for better treatment outcomes because a small subpopulation of CSCs remains resistant to chemotherapy and gives rise to recurrent cancer.
Recently,
TAGLN2 has emerged as a biomarker that plays an essential role in developing various types of cancer [
36]. Alteration of
TAGLN2 expression has been noted in different types of cancer at both the transcriptional and translational levels, and cancer cell proliferation, invasion, and metastasis might be inhibited by suppressing
TAGLN2. Tumorigenesis and tumor development may be correlated with the deregulation of
TAGLN2. Tumor size, clinical stage, histological neural invasion, and lymph node metastasis are closely associated with TAGLN2 in bladder cancer [
37], colorectal cancer [
38,
39], esophageal cancer [
24], and gastric cancer [
40]. Moreover, cancer cell proliferation and EMT, which are responsible for cancer development, dissemination, and resistance to chemotherapy, are decreased by the downregulation of TALGN2 in breast cancer [
41], renal cell carcinoma [
42], cervical cancer [
43], and head and neck squamous cell carcinoma [
22].
Previous studies have suggested that overexpression of TAGLN2 is a potential cause of chemoresistance by increasing EMT properties; however, the mechanism of cancer development and chemoresistance by TAGLN2 has not yet been identified. Recent studies focus on the correlation of TAGLN2 with signaling, including transforming growth factor-beta (TGF-β)/FoxM1 [
25], TGF-β/SMAD4 [
44], IGF1R/phosphoinositide 3-kinase (PI3K)/Akt [
45], and PI3K/phosphatase and tensin homolog (PTEN)/Akt pathways [
41,
46‐
48]. In this study, we demonstrated that the suppression of
TAGLN2 altered not only the expression of EMT-associated proteins, including N-cadherin, Snail, and JAG2, but also those of cMET, AKT, and Nanog in a dose- and time-dependent manner following rTAGLN2 treatment.
Several studies have attempted to overcome chemoresistance by targeting TAGLN2-related signal pathways. SB-T-121,205, a next-generation taxoid, shows antitumor activity by inhibiting the TAGLN2 and PI3K/Akt pathways in human breast cancer cells [
49]. In paclitaxel-resistant breast cancer cells, salvianolic acid A downregulates the expression of TAGLN2 by activating the PI3K/Akt pathway, restoring chemoresistance to paclitaxel [
41]. Here, we inhibited TAGLN2 expression in gemcitabine-resistant BTC cells and restored their chemosensitivity to gemcitabine and other chemotherapeutic drugs, including 5-FU, cisplatin, oxaliplatin, and carboplatin (Fig.
4). Moreover, silencing TAGLN2 in gemcitabine-resistant BTC cells showed both additive and synergistic effects on the therapeutic efficacy of gemcitabine, 5-FU, cisplatin, oxaliplatin, and carboplatin, suggesting the combination therapy of anti-TAGLN2 and a cytotoxic drug as a potential therapeutic strategy to overcome chemoresistance in BTC cells.
In the patient tissues, TAGLN2 is expressed in cancer cells with an overall high expression intensity, making it challenging to analyze its correlation with prognosis. However, it is rarely expressed in normal tissues. TAGLN2 suppression inhibited cancer cell proliferation and reduced resistance to conventional anti-cancer drugs; therefore, TAGLN2 can be applied to most patients with cholangiocarcinoma, suggesting that it is a novel therapeutic target with fewer off-target effects.
Previously, several markers expressed in CAF were reported as prognostic factors associated with the survival of BTC. A previous study has shown that the high expression of IL-33 in both cancer cells and stromal CAFs is associated with better two-year survival of patients with BTC [
50]. The matricellular protein periostin expressed in α-SMA + CAFs was reported as a poor prognostic factor in post-resected BTC [
51]. Another marker, stromal cell-derived factor-1, associated with tumor fibrogenesis and EMT, has also been shown to be correlated with reduced median survival in patients with BTC [
52]. The findings of the present study demonstrated that TAGLN2 expression in patient tissues was increased in stromal tissues around cancer cells and that stromal expression of TAGLN2 was related to patient prognosis. IF analysis of the patient tissues confirmed that the expression of TAGLN2 coincided with that of α-SMA, suggesting that TAGLN2 is expressed in CAF. Furthermore, the IF of CAF derived from patients with cholangiocarcinoma confirmed the expression of TAGLN2 in CAFs. TAGLN2 has been reported as a myCAF marker [
53]; in this study, we report TAGLN2 expression in CAF and its association with CAF from BTC. Even though our findings are based on a few patient samples, increased TAGLN2 expression in CAFs was associated with poor prognosis in patients with BTC. Nevertheless, further studies are required to understand the effect of TAGLN2 on the crosstalk between CAF and cancer cells.
The present study demonstrated that serum TAGLN2 levels significantly increased in patients with cholangiocarcinoma. Compared to CA19-9, the only existing cancer marker, TAGLN2, showed a significantly higher ROC AUC in distinguishing patients with cholangiocarcinoma than those with normal or benign disease (TAGLN2 vs. CA19-9:0.901 vs. 0.799, P = 0.026), moreover, combining with CA19-9, ROC AUC was increased even 0.948. These results suggest that TAGLN2 can be used as a diagnostic marker for cholangiocarcinoma. In particular, 37 of the 44 CA19-9 negative patients (84.1%) showed TAGLN2 elevation above the cutoff, suggesting that TAGLN2 can overcome the low sensitivity of CA19-9. However, because of the unavailability of an efficient TAGLN2 ELISA kit at present, quantitative analysis of TAGLN2 in blood was performed using WB densitometry, which is difficult to apply consistently in various clinical situations. To use TAGLN2 in the blood for diagnosing cholangiocarcinoma, a technical method that can provide more robust and consistent results is needed.
In conclusion, the present study shows that TAGLN2 plays a specific role in tumor proliferation, migration, and invasion and is involved in chemoresistance by inducing EMT-like changes. Targeting TAGLN2 is expected to be a successful anti-cancer therapy for advanced cancer following chemotherapy failure. Further studies to clarify the signaling network and mechanisms of TAGLN2 in carcinogenesis and drug resistance should be conducted to develop new therapeutic approaches for treating chemorefractory BTC.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.