This article is mentioned in:

Introduction

Globally, cancers are considered as one of the highest leading causes of mortality. The pathogenesis of cancer is initiated when normal cells acquire the ability to uncontrollably grow and eventually invade and damage the normal tissues of the body (1). Cancer development is comprised of multiple stages, that is, from precancerous changes to malignant tumors. Furthermore, some cancer cells may spread from their original location to other places in the body via the bloodstream or lymphatic system, referred to as metastasis (2). Embryonic cancer stem cells (CSCs) are known as a small subpopulation of whole cancer cells which exhibit high tumorigenic potential (3,4). In the past, CSCs were simply considered as tumor-initiating cells and also known to be the primary cause for tumorigenesis (5,6). CSCs are known to have additional features including a self-renewal capacity, differentiation into all types of cancers, invasion and metastasis into other tissues and drug resistance (7). These characteristics induce cancer aggressiveness and play a pivotal role in increasing cancer therapy difficulty. Additionally, CSCs have the fundamental properties of all types of cancers including high tumorigenicity, high metastasis potential, immune surveillance evasion, drug resistance and cancer relapse (3,5,6,8,9). Thus, several therapeutic attempts targeting these CSCs have been employed to improve the therapeutic effects and long-term clinical outcome, including efforts to overcome the unique carcinogenic properties of CSCs.

The transcription pluripotency factors, also known as stem cell markers, such as NANOG, SOX2 and OCT4, contribute in maintaining the phenotype of pluripotent embryonic stem (ES) cells in embryonic stem cells (ESCs) and enterochromaffin cells (10–12). Some studies suggest that the abnormal ES cell self-renewal through the hyper-expression of these transcription pluripotency factors promotes carcinogenesis (13–15). Particularly, NANOG overexpression enhances poor prognosis in gastric adenocarcinoma indicating tumor progression (16). Reactive oxygen species (ROS) are known to be a subset of free radicals, which promote tumorigenesis, resulting in oxidative damages of DNA, protein or lipid in normal cells. In the prolonged stress state, cells activate DNA repair pathway, also known as a DNA damage response (DDR), for the repair of damaged DNA and maintaining genome fidelity (17–19). The DDR mechanism is initiated with the induction of phosphorylation in ATM, ATR and DNA-PKcs kinase (20,21). Moreover, Chk1/2 and BRCA1 are phosphorylated, thus activating p53-dependent pathway in the repair response to DNA damage, which leads to the induction of cell cycle arrests, DDR activation and apoptotic mechanism (22).

Cell cycle checkpoints are precisely controlled by both activators and inhibitors to evade normal cell carcinogenesis. Generally, the failed checkpoints for oncogenesis or tumorigenesis allow progression of uncontrolled cell cycle, induce failure of apoptosis and downregulate tumor suppressor factors. Cyclin-dependent kinases, such as CDK4, are considered to be significant cell cycle regulators through positively regulating multiple checkpoints (23,24). However, CDK inhibitors (CKIs) such as p21 (p21Cip1/Waf1) and p53 are known as tumor suppressor factors that inhibit the abnormal cell cycle and carcinogenesis (25–27). Therefore, carcinogenesis in normal cells is related to abnormal operation of CDKs or CKIs, which indicates loss of the ability to induce DNA repair or apoptosis. These cell cycle checkpoints are being studied as target sites for anticancer treatment. Matrix metalloproteinases (MMPs), such as MMP-2, −3 or −9, are knowns as the family of metzincin proteases, which play a role in the degradation of various proteins in the extracellular matrix (ECM) (28,29). In tumor cells, it plays an essential role in a wide range of tumorigenesis, tumor growth, tumor invasion and metastasis (30–32). Additionally, MMPs are emerging as a key player in the entrance or exit of circulating cancer cells in the blood vessels (33). Hence, tumor-associated MMPs are good targets for developing MMP inhibitors for anticancer therapy.

Chemotherapy using various types of drugs has been used to effectively treat several types of cancer. Cancer cells are known to grow faster than normal cells, which indicates that it is easier for drugs to attack these fast-growing cells. However, similar to other cancer treatments, it often causes various side effects, including fatigue, nausea, vomiting, various pains or memory loss (34,35). Additionally, although chemotherapy is initiated depending on the cancer type, location, drugs and health condition of the patient, chronic drug exposure has been linked to serious side effects, including neutropenia, arrhythmia, neuropathy and cardiovascular breakdown. Thus, cancer treatment using natural ingredients could be a more optimal alternative as it has been known to alleviate side effects compared with chemotherapeutic drugs (36–40). Recently, anticancer chemotherapy using natural ingredients has been widely employed to extend the therapeutic domain in chemotherapeutic agents, decrease side effects or drug resistance (41–44).

Natural gallic acid (3,4,5-trihydroxybenzoic acid; GA) is a secondary metabolite, which is widely distributed in natural plants, vegetables, fruits and green tea (45–47). GA is considered to have numerous pharmacological properties, including antibacterial, anti-inflammatory, antioxidant, antiviral and antitumor activities (48–55). Furthermore, it has anticancer properties, including apoptosis induction and suppression of angiogenesis and metastasis. Some studies have reported that GA selectively activates the induction of cancer cell apoptosis in HeLa, HCT-15, SH-SY5Y and NSCLC cells (56–60). In addition, it suppresses cell proliferation and metastasis through the increase or decrease of fatty acid synthase in estrogen receptor (ER) alpha level in TSGH-8301 bladder cancer cells (60). A study suggests that GA decreases the progression of T24 bladder cancer cell via inducing mitochondrial dysfunction and suppressing the PI3K/Akt and NF-κB signaling pathway (61).

The present study revealed that natural GA showed multiple anticancer functions in two NTERA-2 and NCCIT human embryonic carcinoma cells. The anticancer properties of GA in terms of proliferation, apoptosis, cell cycle distribution, migration and invasion were assessed. It was demonstrated that GA effectively suppressed the cancer characteristics of CSC cells. Additionally, it was revealed that the reduction of MMPs by GA was associated with EGFR/JAK2/STAT5 signaling inhibition.

Materials and methods

Cell Culture

Embryonic carcinoma cells (NCCIT and NTERA-2 cells with biomarkers of NANOG, SOX2 and OCT4) were obtained from the Korea Cell Line Bank. NTERA-2 cells were maintained using Dulbecco's Modified Eagle's Medium (DMEM; cat. no. L0103; Biowest) and NCCIT cells were cultured using RPMI-1640 media (cat. no. L0498; Biowest) containing 10% fetal bovine serum (FBS; cat. no. A5670801; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (cat. no. 15140122; Gibco; Thermo Fisher Scientific, Inc.) at 37°C in 5% CO2; 5×105 cells/60 mm dish were cultured up to 70–80% confluence for 48 h and the adherent cells were harvested with trypsin-EDTA solution (cat. no. T4049; MilliporeSigma). The cells were either treated or left untreated with GA and were then further cultured for 24 h.

Reagents and antibodies

GA was bought from MilliporeSigma and dissolved in distilled water (stock concentration of 100 mM). MitoSOX and CM-H2DCFDA were obtained from Invitrogen (Thermo Fisher Scientific, Inc.). The primary antibodies were purchased from Santa Cruz Biotechnology, Inc.: p21 (cat. no. sc-756), p53 (cat. no. sc-126), cyclin E (cat. no. sc-481), MMP9 (cat. no. sc-13520), STAT5b (cat. no. sc-1656), CDK4 (cat. no. sc-260) and β-actin (cat. no. sc-47778). Cell Signaling Technology, Inc. supplied: p27 Kip1 (cat. no. 3686), p-EGFR (cat. no. 3777), EGFR (cat. no. 4267), phosphorylated (p-)JAK2 (cat. no. 3776), JAK2 (cat. no. 3230), p-STAT5 (cat. no. 9351), pCHK1 (cat. no. 2348), p-CHK2 (cat. no. 2197), p-ATM (cat. no. 5883), p-Histone (cat. no. 9718) and p-BRCA1 (cat. no. 9009). MilliporeSigma supplied MMP3 (cat. no. AB2963), OCT4 (cat. no. MABD76), SOX2 (cat. no. MAB4423) and NANOG (cat. no. MABD24) Abcam supplied Cyclin D1 (cat. no. ab6152) and EnoGene MMP2 (cat. no. E90317). Secondary antibodies were purchased from Cell Signaling Technology (cat. nos. 7074; anti-rabbit and 7076; anti-mouse). All antibodies were incubated at 4°C for 12–16 h.

Morphological analysis

NTERA-2 and NCCIT cells were cultured at a density of 1.5×105 cells per well and treated with or without GA for 24 h, followed by washing with phosphate-buffered saline (PBS) twice and again incubated with DAPI staining solution for 1 h. A fluorescence microscope was used to capture the images (Olympus IX71/DP72; Olympus Corporation).

Cell viability assay

MTT (cat. no. M6494; Thermo Fisher Scientific, Inc.) is used to detect reductive metabolism in cells for viability, proliferation and cytotoxicity assays. Cells (3×103 cells per well) were cultured in a 96-well plate for 24 h. Thereafter, they were incubated using dimethyl sulfoxide (DMSO) as vehicle control or using different concentrations of GA (20–400 µM) at 37°C for 24 h. The next day, treatment with 5 mg/ml of MTT reagent was performed and incubation for 4 h at 37°C. After washing with 1X PBS, the formazan product was then dissolved in DMSO. At a 560-nm wavelength, the formazan product absorbance was measured with an Ultra Multifunctional Microplate Reader (Tecan Group, Ltd.).

Apoptosis analysis

An Apoptosis assay kit (cat. no. 556570; BD Biosciences) was used to measure apoptosis in NCCIT and NTERA-2 cells. First, the GA-treated or GA-untreated cells were washed with PBS and resuspended in a binding buffer at a concentration of 1×106 cells. Thereafter, cells were stained with Annexin V-FITC and propidium iodide for 10 min in a dark box at room temperature. Finally, the percentage of apoptotic cells was measured using flow cytometry using a FACSCalibur flow cytometer (BD Biosciences) and dead cells are initially filtered through a ‘singlets’ gate that excludes aggregates, using the height vs. area signals of the same parameter (such as side scatter or forward scatter). Then, cells were gated for detecting apoptotic cells on Annexin V-FITC (FL1) vs. propidium iodide (PI) (FL3). FlowJo v10 software (FlowJo LLC) was used to perform the analysis. The apoptotic rates were calculated with apoptotic cells (Annexin V+/PI-), late apoptotic cells (Annexin V+/PI+), and necrotic cells (Annexin V-/PI+).

Cell cycle analysis

The DNA content from cells with or without GA was determined using a BD Cycletest Plus DNA Reagent Kit (BD Biosciences) according to the manufacturer's protocol. The samples (5×105 cells) were analyzed using a FACSCalibur flow cytometer (BD Biosciences).

Western blot analysis

All protein samples were obtained using a radioimmunoprecipitation assay lysis buffer (cat. no. 20–188; MilliporeSigma) on ice for 20 min and determined by the Bradford assay. The proteins isolated (10 µg/well) were resolved on sodium dodecyl sulfate (SDS)-polyacrylamide gels (6–15%) and transferred to nitrocellulose western blotting membrane (cat. no. 10600002; Cytiva). Then 2% bovine serum albumin (BSA; cat. no. 5217; R&D systems) was used for blocking at 4°C overnight. Target proteins were identified using immunoblotting assay at 4°C overnight. The primary antibodies (1:1,000 dilution) and secondary antibodies (1:5,000-10,000) used are in the Reagents and antibodies section. The iBright™ CL1500 Imaging System (cat. no. A44114; Invitrogen; Thermo Fisher Scientific, Inc.) was used to capture images and analyze data from immunoblotting assay.

Reverse transcription-quantitative (RT-q) PCR

At between 70–80% confluence, cells were gently washed twice with PBS. Using TRIzol® (cat. no. 15596018; Thermo Fisher Scientific, Inc.), total RNAs were isolated and quantified by Nanodrop (ND-1000; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Subsequently, a thermal cycler was used to prepare cDNA from the total RNA using a first-strand cDNA synthesis kit (Bioneer Corporation) for RT-qPCR according to the manufacturer's instructions. The qPCR was performed using a LightCycler 480II (Roche Diagnostics). It was used for qPCR as follows: 2 µl diluted cDNA was mixed with 10 µl TB Green Advantage Premix (cat. no. 639676; Takara Bio, Inc.) and 1 µl (100 pM) each of forward and reverse primers according to the manufacturer's instructions. The cycling conditions were: 95°C for 5 min for the initial denaturation, followed by 40 cycles of 95°C for 40 sec, 58°C for 40 sec, 72°C for 40 sec and, finally, extension for 5 min at 72°C. All reactions were conducted three times and normalized to GAPDH and quantifications were conducted using obtained Cp values. The sequences of all primers with GAPDH are in Table I.

Table I. Primer sequences for reverse transcription-quantitative PCR.

Table I.

Primer sequences for reverse transcription-quantitative PCR.

CDK4	Sense	5′-CCCGAAGTTCTTCTGCAGTC-3′
	Antisense	5′-CTGGTCGGCTTCAGAGTTTC-3′
Cyclin D1	Sense	5′-TGTTTGCAAGCAGGACTTTG-3′
	Antisense	5′-TCATCCTGGCAATGTGAGAA-3′
Cyclin E	Sense	5′-ATCCTCCAAAGTTGCACCAG-3′
	Antisense	5′-AGGGGACTTAAACGCCACTT-3′
p21	Sense	5′-ATGAAATTCACCCCCTTTCC-3′
	Antisense	5′-AGGTGAGGGGACTCCAAAGT-3′
p27	Sense	5′-CCGGCTAACTCTGAGGACAC-3′
	Antisense	5′-TTGCAGGTCGCTTCCTTATT-3′
SOX2	Sense	5′-CTGCAGTACAACTCCATGAC-3′
	Antisense	5′-GAGTGGGAGGAAGAGGTAAC-3′
Nanog	Sense	5′-ACCAGTCCCAAAGGCAAACA-3′
	Antisense	5′-TCTGCTGGAGGCTGAGGTAT-3′
Oct4	Sense	5′-CAAAGCAGAAACCCTCGTGC-3′
	Antisense	5′-AACCACACTCGGACCACATC-3′
MMP2	Sense	5′-TGATGGCATCGCTCAGATCC-3′
	Antisense	5′-GGCCTCGTATACCGCATCAA-3′
MMP3	Sense	5′-CACAGACCTGACTCGGTTCC-3′
	Antisense	5′-AGGTTCTGGAGGGACAGGTT-3′
MMP9	Sense	5′-GGACAAGCTCTTCGGCTTCT-3′
	Antisense	5′-TCGCTGGTACAGGTCGAGTA-3′
GAPDH	Sense	5′-CCCACTCCTCCACCTTTGAC-3′
	Antisense	5′-TCCTCTTGTGCTCTTGCTGG-3′

Tumorsphere formation assay

Cells were cultured in DMEM/F-12-containing growth supplements, epidermal growth factor, basic fibroblast growth factor and B27 in low attachment six-well plates together with or without GA for 14 days. Cell status was checked on days 0, 7 or 14 using an optical microscope (BX-51; Olympus Corporation) at 100× and 200× magnification, and cells harvested for immunoblotting assay.

Mitochondrial and cellular ROS Analysis

1×106 cells were stained with 5 µM of MitoSOX to evaluate mROS or 5 µM of CM-H2DCFDA to evaluate cellular ROS for 20 min at 37°C. Then, the stained cells were used for ROS analysis using a FACSCalibur flow cytometer (BD Biosciences). FlowJo v10 software (FlowJo LLC) was used for the analysis.

Comet assay

Cellular DNA damage was measured using the comet assay kit (cat. no. 238544; Abcam) following the manufacturer's protocol. This assay is a single-cell gel electrophoresis method for a simple evaluation of cellular DNA damage. First, a base layer of comet agarose was created on a slide and then a layer of cells and agarose was added, followed by lysis. Next, under neutral conditions, electrophoresis was performed and cells were stained using a DNA dye. Finally, a fluorescence microscope was used to observe cell morphology (Olympus IX71/DP72; Olympus Corporation).

Invasion assay

Matrigel was used to pre-coat plates overnight at 4°C. For the invasion assay, 5×104 cells with or without GA were inserted into each invasion chamber (BD Biocoat; BD Biosciences) and further incubated at 37°C for 24 h. Then, cells were stained with crystal violet at room temperate for 30 min and the invaded cells were observed using an optical microscope (BX-51; Olympus Corporation) at 100× and 200× magnification and counted.

Wound healing assay

NTERA-2 and NCCIT cells (1×105 per well) were seeded and incubated in six-well plates. After a confluent monolayer was attained, the cell layers were scratched using a 1,000 µl pipette tip and immediately incubated with GA (0 or 200 µM) for 24 h. Images were captured at different time intervals to assess the wound edges through an optical microscope (BX-51; Olympus Corporation) at 100× and 200× magnification and closure of the wound was quantified by ImageJ (v1.51; National Institutes of Health). All cells were cultured with culture medium without FBS.

Chromatin immunoprecipitation (ChIP) assay. A ChIP assay kit was purchased from MilliporeSigma (cat. no. 17–295) and the ChIP assay was performed according to the manufacturer's protocol. Cells (1×106) were fixed in 1% formaldehyde and quenched with 1.25 M glycine, washed with ice-cold PBS, suspended in a nuclear preparation and shearing buffer, and then sonicated (on/off 20 sec during 20 min on ice). The sheared DNA was centrifuged at 13,000 × g for 15 min at 4°C, followed by protein/DNA immunoprecipitation from the cleared supernatant as follows. The clarified supernatant was diluted with buffer (1:1 ratio), and 5 µl aliquots of the diluted samples were used as internal controls. Next, the diluted supernatant was incubated with STAT5b anti-bodies in pre-coated wells for 90 min at room temperature. The controls were incubated with normal goat-IgG and anti-RNA polymerase II. Unbound DNA was removed and bound DNA collected using the cross-link reversal method with DNA release buffer containing proteinase K. The released DNA and internal control DNA were purified using the GenElute Binding Column G (cat. no. 17-295; MilliporeSigma). DNAs from each cell were quantified using a PCR instrument with the following primers for MMP2: Sense; 5′-TGATGGCATCGCTCAGATCC-3′ and Antisense; 5′-GGCCTCGTATACCGCATCAA-3′.

Statistical analyses

All experiments were performed in triplicate. The values of three independent experiments conducted in triplicates (n=3) were represented as mean ± standard error of the mean. The controls were set to 100. One way analysis of variance (ANOVA) or Student's t-test were used for the statistical analyses. One-way ANOVA was also performed using Tukey's post hoc test. The analyses were performed using the SAS 9.3 software (SAS Institute, Inc.). The relative density of proteins by the effects of GA was compared with the GA non-treated control. P<0.05 was considered to indicate a statistically significant difference.

Results

GA inhibits the propagation of embryonic CSCs and induces apoptosis

The structure of GA, also known as 3,4,5-trihydroxy benzoic acid, is shown by Fig. 1A. First, the present study performed the MTT assay for confirming the inhibition ability of cell proliferation by GA in embryonic CSCs. It confirmed a dose-dependent inhibition of cell viability through GA treatment in CSCs (Fig. 1B). The present study used 200 µM GA as the IC50 dose and 100- or 200 µM GA concentrations for dose-based studies. Then, the status of embryonic CSCs following 200 µM GA treatment was estimated using DAPI staining. This confirmed the characteristic features of apoptosis, such as the nuclear shrinkage and fragmentation as well as the chromatin condensation in CSCs treated with GA. A decrease in cell number following GA treatment was clearly observed compared with non-treated control (Fig. 1C). The inhibition of CSCs growth via GA treatment was verified with bright-field microscopic analysis. These results may provide evidence that GA can induce apoptosis. FACS-based Annexin V/PI staining assay was applied to assess the effects of GA on the apoptosis and necrosis of NTERA-2 and NCCIT cells. The assay may show that GA induced the increase in early-stage apoptosis (Annexin V+/PI-cells), late-stage apoptosis (Annexin V+/PI+ cells) and necrosis (Annexin V-/PI+ cells). Treatment with 200-µM GA into CSCs demonstrated the inhibitory effect of GA, suggesting that GA may induce apoptosis. In addition, increased cell death by 200-µM GA was observed and results revealed the induction of cell apoptosis by GA in both NTERA-2 and NCCIT cells (Fig. 1D).

Figure 1. GA inhibits cell viability and induce apoptosis in CSCs. (A) Chemical structure of gallic acid. (B) The graphs display the cytotoxicity assay results in embryonic CSCs at 24 h. (C) GA induced nuclear abnormality in embryonic CSCs (scale bar, 200 µm). (D) The flow cytometry analysis results demonstrated induction of late apoptosis in embryonic CSCs with 0 or 200 µM GA, respectively, for 24 h. ***P<0.001 (ANOVA test) and #P<0.001 vs. control. GA, gallic acid; CSCs, cancer stem cells.

GA induces cell cycle arrest in embryonic CSCs

The present study used 100- or 200-µM GA for further studies. Previous results suggested that GA may induce cell cycle arrest. The present study verified that cell cycle arrest in G0/G1 phase, which was increased through the administration of 200 µM GA, was found in CSCs through using a flow cytometry (Fig. 2A). To confirm the inhibitory induction of cell cycle by GA treatment, western blotting assay was performed to identify the lower-expression of cell cycle markers (CDK4 or cyclin D1 and E) and hyper-expression of tumor suppressor proteins (p21, p27 or p53) was confirmed (Fig. 2B). To confirm the RNA levels, qPCR was performed to assess for the expression of markers (p21, p27, CDK4 or cyclin D and E) and these results clearly demonstrated the inhibition of cell cycle arrest by GA (Fig. 2C). These results suggest anticancer activity of GA against CSCs.

Figure 2. GA enhances the arrest of G0/G1 cell cycle in CSCs. (A) The flow cytometry results demonstrated the cell cycle distribution with 0 or 200 µM GA, respectively, for 24 h. Graphics display a G0/G1 arrest by GA in embryonic CSCs. (B) Immunoblotting analysis showed the expression patterns of p21, p27, p53, CDK4 or cyclin D1 and E with 0, 100, or 200 µM GA, respectively, for 24 h. All proteins were normalized to β-actin levels. (C) Reverse transcription-quantitative PCR represented the genetic expression of cell cycle factors, including CDKN1A, CDKN1B, CDK4, CCND1 or CCNE1 mRNA. ***P<0.001 (ANOVA test) and #P<0.001 vs. control. GA, gallic acid; CSCs, cancer stem cells.

GA suppresses cancer stemness through the inhibition of tumorsphere formation in embryonic CSCs

For an improved replication of the conditions in vivo, tumorsphere formation assay was performed for 3D culture. It was determined whether GA could downregulate the expression of CSC markers. In the analysis of the CSC markers, NANOG, OCT4 or SOX2, in NTERA-2 and NCCIT cells upon GA treatment, a decrease in the expression level of CSC markers was clearly observed (Fig. 3A). Using RT-qPCR, CSC marker repression by GA was also confirmed at the level of mRNA transcripts (Fig. 3B). These results indicated inhibition of proliferation of CSCs by GA. A tumorsphere formation assay was further performed using NTERA-2 and NCCIT cells to evaluate the GA ability to inhibit the two CSCs. NTERA-2 and NCCIT cells were cultured in the tumorsphere medium using low attachment plates with GA for 14-days. Images were captured using a microscope, which indicated a significant reduction in tumorspheres following the treatment of 200 µM GA compared with untreated controls (Fig. 3C). To confirm CSCs suppression by GA on the protein level using these samples, western blotting was used to identify the expression levels of specific CSC markers after sphere formation culture for 14 days. A significant downregulation of stem cell markers, NANOG, OCT4 or SOX2, by GA was observed (Fig. 3D). These results demonstrated the repression of target CSCs by GA.

Figure 3. GA inhibits cancer stemness in CSCs. (A) The expression patterns of CSC markers, NANOG, SOX2 or OCT4, were incubated with 0 or 200 µM GA, respectively, for 24 h. (B) Reverse transcription-quantitative PCR of CSC marker mRNAs was performed and normalized to GAPDH mRNA. ***P<0.001 (ANOVA test) and #P<0.001 vs. control. (C) The tumorsphere formation assay of CSCs were performed with 0 or 200 µM GA, respectively, for 14 days (scale bar, 200 µm). (D) After 14 days, the CSC marker expression levels of NANOG, SOX2 or OCT4 were measured via western blotting. GA, gallic acid; CSCs, cancer stem cells.

GA activates ROS mediated-DNA damage in embryonic CSCs

The present study identified GA activity that induces cell cycle arrest and apoptosis in CSCs. Thus, it further confirmed whether ROS was also induced by GA, as this is crucial in anticancer study or application. MitoSOX and CM-H2DCFDA reagents are widely used to identify mitochondrial and cellular ROS. MitoSOX and CM-H2DCFDA reagents were used for detecting the mitochondrial and cellular ROS by GA. It was found that the treatment of 200-µM GA can show the considerable induction of cellular ROS levels in both NTERA-2 and NCCIT cells (Fig. 4A). As the mitochondria are known as the major source of cellular ROS in cells, an increase in cellular ROS levels by GA may imply ROS induction in mitochondria. The induction of higher mitochondrial ROS as in the case of cellular ROS by 200-µM GA in CSCs was also detected by flow cytometry analysis (Fig. 4B). Based on the hypothesis that the induction of cell death, cell cycle arrest or apoptosis was a result of DNA damage, presence of DNA damage was therefore confirmed. Excessive ROS product is known to cause DNA damage and induce the DDR pathway. To confirm this, the induction of DDR protein phosphorylation was investigated and confirmed upregulation in DDR proteins, p-BRCA1, ATM, CHK1, CHK2 or histone when treated with increased GA concentrations, focusing on NTERA-2 and NCCIT cells (Fig. 4C). Furthermore, the activity of GA to promote DDR mechanism was evaluated using a comet assay for detecting DNA fragmentation. A significant increase in comet-positive cells was also noted, including the length of comet in GA treated-cells compared with untreated-control cells (Fig. 4D). These results may validate the capability of GA to cause DNA damage in embryonic CSCs.

Figure 4. GA enhances ROS generation and DNA damage induction in CSCs. (A) The flow cytometry analysis for detecting cellular ROS with 0 or 200 µM GA, respectively, for 24 h. The image showed the embryonic CSCs with induction of cellular ROS. (B) The flow cytometry analysis for detecting mitochondrial ROS with 0 or 200 µM GA, respectively, for 24 h. The graphical image revealed embryonic CSCs with the induction of mitochondrial ROS. (C) The western blotting analysis results showed the induction of the DDR factors, p-BRCA1, p-ATM, p-Chk1, pChk2 and p-Histone with 0, 100 or 200 µM GA, respectively, for 24 h. (D) The comet assay results indicated the fragmented DNA migration from the nucleoid body and the percentage of comet-positive CSCs with 0 or 200 µM GA, respectively, for 24 h. ***P<0.001 (ANOVA test) vs. control. GA, gallic acid; ROS, reactive oxygen species; CSCs, cancer stem cells; p-, phosphorylated.

GA inhibits migration and invasion of embryonic CSCs

GA prevented CSCs from normally expanding and growing through DNA damage and inhibition of sphere culture growth. Hence, it was hypothesized that GA can inhibit tumor metastasis and invasion. The ability of GA to suppress CSCs migration and invasion was analyzed to confirm metastasis. First, the present study analyzed for molecular markers to confirm the capacity for invasion and the decrease of MMP proteins, MMP-2, −3 or −9, by GA was identified, suggesting the reduction of CSC progression and metastasis (Fig. 5A). These results were then validated via the analysis of gene expression patterns at the level of mRNA transcripts and GA induced the downregulated mRNA levels of MMPs in two CSCs (Fig. 5B). An invasion assay was also performed to confirm whether GA can induce the inhibition of CSCs metastasis and the results indicated a significant suppression of invaded cells in GA treated-CSCs (Fig. 5C). Moreover, to determine whether CSCs can migrate after exposure to GA, migration-inhibiting effects by GA were confirmed through a wound-healing assay (Fig. 5D). Overall, these results indicated that GA effectively repressed cell migration and invasion through MMPs expression inhibition in embryonic CSCs.

Figure 5. GA weakens migration and invasion in embryonic CSCs. (A) Immunoblotting analysis reveals the expression patterns of MMP-2, −3 or −9 protein with 0, 100 or 200 µM GA, respectively, for 24 h. (B) The analysis of RT-qPCR indicates the genetic expression of cell cycle factors including MMP-2, −3 or −9 mRNA. (C) The results of a Matrigel invasion assay indicate the suppression of embryonic CSCs invasion with 0 or 200 µM GA, respectively, for 24 h (scale bar, 100 µm). (D) The wound healing assay shows the inhibitory migration of embryonic CSC with 0 or 200 µM GA, respectively, for 24 h (scale bar, 200 µm). ***P<0.001, **P<0.01 and *P<0.05 (ANOVA test); #P<0.001 vs. control. GA, gallic acid; CSCs, cancer stem cells; MMP, matrix metalloproteinase.

GA inhibits the EGFR-STAT5 pathway and DNA-protein binding activity by STAT5

After confirmation that GA can inhibit migration and invasion, the mechanism of molecular signaling by GA was evaluated. The EGFR/JAK2/STAT5b pathway is known to be associated with tumor invasion and migration. To validate these results, the EGFR-STAT pathway was assessed through GA treatment and the suppression of EGFR phosphorylation blocked cellular signaling toward the downstream targets of EGFR (Fig. 6A). Additionally, to assess the relationship between STAT5 and MMP-2, GA and/or a specific STAT5 siRNA was used to confirm the STAT5-mediated MMP-2 expression (Fig. 6B). These results showed a significant inhibition of p-STAT5 and MMP-2 expression in GA-treated samples and its combination with STAT5 siRNA samples (Fig. 6B). To demonstrate the p-STAT5 binding onto the MMP-2 promoter region by GA inhibition, primers specific to the MMP-2 genome were designed and its binding was confirmed using ChIP assay. GA-treated samples indicated the considerable suppression of STAT5/MMP-2 binding (Fig. 6C). This result clearly suggested the role of the EGFR/STAT5/MMP-2 signaling cascade in the anticancer ability of GA against embryonic CSCs. Altogether, GA can inhibit cancer hallmarks against NCCIT and NTERA-2 human embryonic carcinoma cells through EGFR-mediated JAK2/STAT5 signaling mechanisms (Fig. 7).

Figure 6. GA inhibits migration and invasion through the decrease of the EGFR-STAT5 pathway in embryonic CSCs. (A) The western blotting assay in embryonic CSCs with 0, 100 or 200 µM GA, respectively, for 24 h for p-EGFR, JAK2 or STAT5. (B) Immunoblotting analysis showed that treatment with 30 pM STAT5 siRNA or 200 µM GA for 24 h demonstrated the decrease of MMP-2, STAT5 or p-STAT5b levels. (C) The ChIP assay showed that GA inhibited the formation of the STAT5b/MMP2 complexes in embryonic CSCs with 200 µM GA, respectively, for 24 h. ***P<0.001 (Student's t-test). GA, gallic acid; CSCs, cancer stem cells; p-, phosphorylated; MMP, matrix metalloproteinase.

Figure 7. Molecular regulatory mechanism shows the anticancer activities of gallic acid in embryonic NCCIT and NTERA-2 cells. ROS, reactive oxygen species; p-, phosphorylization; EGFR, epidermal growth factor receptor; EGF, epidermal growth factor; JAK2, Janus Kinase 2; STAT5b, signal transducer and activator of transcription 5b.

Discussion

CSCs are considered to be a subpopulation of tumor cells, commonly observed in a majority of liquid and solid cancers (3–6). CSCs feature specific characteristics, including self-renewal, tumorigenesis, multilineage differentiation potential and tumorigenicity (7,62). Success in cancer treatment using drugs is mainly dependent not only on how it affects targeted cancer cells but also on its ability to prevent cancer recurrence of CSCs. Although several types of chemotherapeutic drugs successfully inhibit tumor progression, in a number of cases these fail to target CSCs which ultimately results in cancer recurrence (63,64). CSCs possess a formidable ability to evade or resist the killing effect of chemotherapeutic drugs through genetic instability, hypoxic stability and expression of drug efflux transports and antiapoptotic proteins (65–68). It has been shown that various types of natural compounds have anticancer effects, indicating their potential as a drug (11,36,69). It has also been shown that a chemotherapeutic drug and a natural compound exhibit optimal efficacy when used in combination compared with using the drug alone (70). Therefore, chemotherapy using natural compounds may be an attractive method as it reduces side effects in both cancer cells and CSCs.

GA is a natural phenolic compound, well-recognized to have anticancer ability against diverse types of cancer cells in vitro and in vivo (48,56–58,71). Some studies have shown that GA can suppress cell proliferation by inducing cell cycle arrest and apoptosis in lung and ovarian cancer cells (72,73). The present study revealed that GA induced late apoptosis in both NTERA-2 and NCCIT cells compared with others (that is, early apoptosis and necrosis). Furthermore, GA induced the arrest of G0/G1 cell cycle through the downregulation of CDK4 or cyclin D1 and E and upregulation of p21, p27 or p53, suggesting its anticancer activity against CSCs.

CSC markers play a role in transcription factors, essential in maintaining the pluripotent ESC phenotype in CSCs (10). Furthermore, they are important factors in forming CSC tumorspheres. Thus, CSC markers are deemed good targets for drug development. In previous studies, a natural compound exhibited inhibition of sphere formation in CSCs (11) and suppressed the expression of Wnt/β-catenin-mediated CSC markers (74). The present study found that GA inhibited cancer stemness through the suppression of tumorsphere formation in CSCs. GA effectively downregulated CSC markers in mRNA and protein levels and inhibited CSC marker-mediated tumorsphere formation. Generally, prolonged ROS generation is known to induce to DNA fragmentation, genomic instability and loss of heterozygosity. ROS play a pivotal role in tumorigenesis by inducing DNA damage that results in DDR mechanism. A number of natural compounds can activate the induction of DDR mechanism in cancer cells, finally leading to cell cycle arrest or apoptosis, suggesting that they can be possible candidates for further studies. Prior studies have demonstrated that GA prompts the induction of DNA damage and the inhibition of DNA repair-associated protein expression, resulting in DDR mechanism in some cancer cells (75–77). The present study observed that GA elevated cellular ROS and mitochondrial ROS in CSCs, suggesting an anticancer activity of GA. Furthermore, a high ROS generation by GA upregulated the DDR mechanism proteins, such as BRCA1, ATM or CHK1 kinase which are central sensors or regulator in DDR in cells. The results indicated that the increase in the expression of these kinases by GA enhanced the arrest of G0/G1 cell cycle and apoptosis in CSCs. MMPs perform a key role in the decrease of various proteins in the ECM, resulting in tumorigenesis, tumor growth and tumor invasion and metastasis (30,31). Although the PI3K/AKT/mTOR pathway is related to tumor cell invasion and migration, the JAK/STAT pathway regulates the expression of MMPs, playing a crucial role in cancer progression via invasion or angiogenesis (33,78). Its regulation could be considered as one of the potential targets for halting tumor invasion or migration and thus cancer progression. Our previous study shows that a natural compound markedly inhibits cell invasion in compound-treated cancer cell (39). Thus, the present study demonstrated the decrease of these transmembrane signals that finally control MMPs in CSCs through GA. Eventually, GA decreased the EGFR/STAT5 pathway that enhanced MMP expression, leading to a significant inhibition of the STAT5/MMP2 binding in the nucleus. The present study confirmed that GA has anti-cell invasion and anti-migration properties by inhibiting the JAK/STAT pathway. However, it is necessary to additionally conform that GA suppresses invasion and migration of CSCs by inhibiting PI3K/AKT/mTOR pathway.

The present study revealed that GA targets embryonic CSCs through the inhibition of cell proliferation. In addition, it was also evident that GA enhanced the induction of cellular and mitochondrial ROS, leading to DDR against to CSCs. In addition, it revealed that GA resulted in late apoptosis via the arrest of G0/G1 cell cycle. Finally, the significant suppression of MMPs, which are key factors for tumor invasion, was demonstrated by GA. Altogether, GA could be an attractive agent for preventing cancer recurrence by targeting CSCs.

Acknowledgements

Not applicable.

Funding

The present study was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) to KJ. (grant no. RS-2024-00450676). It was also supported by the Basic Science Research Program to Research Institute for Basic Sciences (RIBS) of Jeju National University through the National Research Foundation of Korea (NRF) funded by the Ministry of Education to SB. (grant no. 2019R1A6A1A10072987) and the Korea Basic Science Institute (National research Facilities and Equipment Center) grant funded by the Ministry of Education. (grant no. 2023R1A6C101A045). The present study was also supported by the faculty research fund of Sejong University in 2024.

Availability of data and materials

The data generated in the present study are included in the figures and/or tables of this article.

Authors' contributions

KJ designed the experiments and wrote the manuscript. DK and SB performed all the experiments and analyzed the data. DK, SB and KJ confirm the authenticity of all the raw data and helped to revise the manuscript. All authros read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References