Pantoprazole Inhibited Metastasis and Angiogenesis Through Wnt/β-Catenin Signaling Pathway in Gastric Cancer Stem-Like Cells

Abstract

Background:

The research on different therapy approaches is critical to find an efficient gastric cancer treatment. The processes of new drug production are very expensive and also take about 15 years. Therefore, the use of existing drugs for treating new diseases may be beneficial in pharmaceutical biotechnology. Thus, we investigated the influence of pantoprazole on Wnt signaling pathway, metastasis, stemness markers, and in vitro tube formation in gastric cancer stem-like cells (GCSCs).

Materials and Methods:

GCSCs were isolated from MKN-45 cells on a nonadhesive surface. Cell viability, angiogenesis, metastasis, and transcription of CTNNB1, WNT1, SMARCD1, CTNNBIP1, KREMEN1, and SUFU genes and protein levels of CTNNBIP1, SMARCD1, and KREMEN1 were measured by trypan blue, tube formation assays, zymography, the real-time RT-PCR, and Western blotting, respectively.

Results:

Our findings represented that pantoprazole decreased the cell viability, angiogenesis, metastasis, and stemness features of GCSCs. Also, pantoprazole had an inhibitory impact on Wnt signaling pathway by modulating the transcription level of CTNNB1, WNT1, SMARCD1, CTNNBIP1, KREMEN1, and SUFU genes and protein level of CTNNBIP1, SMARCD1, and KREMEN1.

Conclusions:

We showed that pantoprazole may reduce the tumorigenicity of GCSCs through the Wnt signaling pathway. Therefore, pantoprazole may be an assistance treatment for gastric cancer therapy.

INTRODUCTION

The fifth most commonly diagnosed cancer is gastric cancer (GC) in the world, and about 660,000 deaths yearly makes GC fifth cancer leading to death.[1] Asian countries especially Eastern Asian countries such as Mongolia, Japan, China, and the Republic of Korea have the highest rates of GC. The incidence of GC in men especially in Western Asian countries such as Iran and Turkmenistan is 2-fold higher than in women.[2] In the past decade, the incidence and mortality rate of GC have been decreased in most countries, but in Iran, the incidence rate of GC has been increased.[3] Gastric cancer is a complex disease, and dysregulation of several cell signaling pathways is involved in the pathogenesis of GC.[4] Unfortunately, there is no effective treatment to inhibit the oncogenic pathways and prevent the progression of GC. Therefore, research on different therapeutic approaches is critical to find an efficient therapy for GC.[5]

Cancer stem cells (CSCs) are a unique and morphologically distinct group of tumor cells with self-renewal capacity and required to generate heterogeneous and variable differentiated tumor cells.[6] For the first time, CSCs were detected in acute myeloid leukemia, and then, CSCs were identified in other solid tumors, including the brain, colon, melanoma, pancreatic cancer, prostate, ovarian, breast, and GC.[7] Almost all tumors recurred after the current cancer therapy strategies because of proliferation of the undifferentiated tumor cells named CSCs.[7] Numerous studies indicated that CSCs were responsible for metastasis, drug resistance, and recurrence of the tumors after resection. Therefore, a critical factor in the treatment of cancers is getting rid of CSCs.[8]

The Wnt signaling pathway has essential functions in many biological processes, including embryogenesis, cell proliferation, apoptosis, and tissue homeostasis.[9] Wnt proteins (19 ligands) bind to their transmembrane cellular receptors (10 members of Frizzled family) and activate canonical and noncanonical signaling.[10] Evidence showed Wnt signaling pathway involved in tumor growth, metastasis, and tumorigenicity of many cancers.[9] Therefore, inhibition of Wnt signaling pathway can be a major research in cancer treatment. Also, several studies indicated that pantoprazole inhibited cell proliferation and metastasis in gastric cancer cell lines through inhibition of Wnt signaling pathway.[11,12]

Research on the production of a new drug and launch process in traditional strategies is very expensive and takes a long time about 10-15 years. Therefore, the use of old drugs for new diseases is a practical field of drug development known as drug repositioning, which is low cost and riskless and reduces time to market the drug.[13] The development of a new drug in traditional strategies costs about $12 billion, whereas the cost of development of a drug is $1.6 billion using a drug repositioning strategy.[8] All clinical safety of an approved drug has been confirmed during Phase I, Phase II, and Phase III, so drug repositioning is a high-reward and low-risk strategy.[13] Pantoprazole, a proton-pump inhibitor (PPI), is an approved drug that irreversibly inhibits the H+/K+ ATP pumps and is used for the treatment of erosive esophagitis and Zollinger–Ellison syndrome and eradication of Helicobacter pylori bacteria and peptic ulcer.[14] In the current study, we investigated the cell and molecular behavior of pantoprazole on Wnt cell signaling, metastasis, and angiogenesis in gastric cancer stem-like cells (GCSCs) derived from a gastric cancer cell line (MKN-45). Also, the effect of pantoprazole on expression of stemness marker genes was identified in GCSCs.

MATERIALS AND METHODS

Gastric cancer cell culture

MKN-45 cells were purchased from the National Cell Bank of Iran affiliated with Pasteur Institute of Iran. The cells were cultured in Dulbecco modified Eagle medium/F12 (DMEM/F12) (Gibco, Belgium) and 10% (v/v) heat-inactivated (FBS) fetal bovine serum (Gibco, Belgium), 100-U/mL penicillin G (Sigma-Aldrich, USA), and 100-µg/mL streptomycin (Sigma-Aldrich, USA) and incubated at 37°C in a humidified CO₂ incubator and 5% CO₂ (Memmert, Germany). The media of cell culture were replaced twice a week.

Separation of GCSCs from gastric cancer cell line

To isolate GCSCs from gastric cancer cell line (MKN-45), the upper layer of dish was covered by a skinny layer of agarose (Merck, Germany). Then, 3 × 10⁴ MKN-45 cells were segregated to single cells and placed as suspended cells in DMEM/F12 and 10% FBS, on the coated surface of dish for 14-16 days. The media of cell culture were carefully replaced with new media every 2 days to eliminate apoptotic and dead cells and only saved colonies of GCSCs.[6,15]

Trypan blue assay

The effect of pantoprazole with 98% purity (Sigma-Aldrich, USA) on cell viability of GCSCs taken from MKN-45 was determined using the Trypan blue assays. First, GCSCs were cultured in triplicate with 3 × 10⁴ cells per well on surface of 24-well plates, which covered by a skinny layer of agarose. Next, GCSCs were treated with pantoprazole in various concentrations (0, 20, 40, 60, 80 and 100 µmol) for a further 24 h. After that, they were broken apart by 0.25% Trypsin–EDTA (Sigma-Aldrich, USA), and cell viability of GCSCs was evaluated by Trypan blue staining and counting of viable cells (white) using a hemocytometer. The percentage of viability was measured as 100% minus the cytotoxicity cell (blue) percentage. All experiments were performed three times.

The extraction of RNA and real-time RT-PCR

The extraction of total RNA from GCSCs was performed by the RNeasy Plus Mini kit (Qiagen, USA) based on the manufacturer’s booklet. Complementary DNA (cDNA) was synthesized as in the PrimeScript™ RT Reagent Kit (TAKARA BIO INC., Japan) manufacturer’s booklet. Primer sequences for Oct3/4, SOX2, NANOG, KLF4, KREMEN1, SUFU, SMARCD1, CTNNBP1, CTNNBIP1, GAPDH, and WNT1 were taken from our previous investigation,[15] but p53 primers were taken from Węglarz et al. [Table 1].[16] Quantitative real-time RT-PCR experiments were performed by SYBR Premix Ex Taq II (TAKARA BIO INC., Japan) in Rotor-Gene 3000 System (Corbett Research, Australia). The results of real-time RT-PCR were analyzed according to measuring the threshold cycle (Ct) values for GAPDH as an endogenous control gene and target genes by the 2^−ΔΔCt formula. The experiments were performed in triplicate.

Table 1.

Primer sequences used for the quantitative gene expression analysis by real-time RT-PCR

Genes Name	Primers	Sequences (5’→3’)	Primers’ Size (bp)	Annealing (°C)	Product size (bp)
GAPDH	Sense	ACTCTGGTAAAGTGGATATTGTTGC	25	54	162
	Antisense	GGAAGATGGTGATGGGATTTC	21
Oct3/4	Sense	GCTTCAGGGTTTCATCCA	18	54	169
	Antisense	GGCGGCAATCATCCTCTG	18
CD44	Sense	TAACAGTTCCTGCATGGGCGGC	20	53	129
	Antisense	CGTGCAAATTCACCAGAAGGC	21
SOX2	Sense	CAACATCACAGAGGAAGTAGACTG	24	54	115
	Antisense	CCTTGGCATGAGATGCAGG	19
NANOG	Sense	AACTCTCCAACATCCTGAACC	21	59	167
	Antisense	GTGGTAGGAAGAGTAAAGGCTG	22
KLF1	Sense	GAACCCACACAGGTGAGAAAC	21	59	171
	Antisense	TGTGTAAGGCGAGGTGGTC	19
WNT1	Sense	CGATGGTGGGGTATTGTGAAC	22	57	112
	Antisense	CCGGATTTTGGCGTATCAGAC	22
CTNNB1	Sense	CAATCCCTGAACTGACAAAACTG	23	58	168
	Antisense	ACGTACAATAGCAGACACCATC	22
CTNNBIP1	Sense	AGACTTGACAACGGTGACAG	20	58	100
	Antisense	AATTAACTTCAGGCAAACAGGTG	23
KREMEN1	Sense	ACAGTCTGAAATACCCCAACG	21	58	185
	Antisense	GTTTCCATGATCCTTGTAGCAG	22
SMARCD1	Sense	CAGGGACCTCAAGACAATGACT	22	59	112
	Antisense	GGAGTAGAAGTATCGGCACACA	22
SUFU	Sense	TGTTGACCGAAGAGTTTGTAGAG	23	58	152
	Antisense	GTGTAGCGGACTGTCGAAC	19

Matrix metalloproteinase (MMP) activity assay

Zymography was used to study MMP activities. Treated GCSCs with pantoprazole in 50-µmol concentration for 24 h and untreated GCSCs were used for gelatinase matrix metalloproteinase (MMP-2 and MMP-9) activity assay. The supernatant of the treated and untreated GCSCs culture were electrophoresed on a polyacrylamide gel (8%) with 1 mg/mL gelatin (Merck, Germany). Then, the gel was submerged in an enzyme-renaturing buffer composed of 2.5% Triton X-100 (Merck, Germany) in 50-mM Tris-HCl pH 7.4 (Merck, Germany), 5-mM CaCl₂ (Merck, Germany), and 1-mM ZnCl₂ (Merck, Germany) on a rotating shaker for 1 h. Next, the gel was placed in developing buffer, including 50-mM Tris, 5-mM CaCl₂, 1-mM ZnCl₂, and pH 7.4 at 37°C. After 24 h, the gel was stained with 0.1% Coomassie Blue R-250 (Merck, Germany) in 40% ethanol and 10% acetic acid, and then, gel was destained by destaining buffer (40% ethanol and 10% acetic acid) to appear the MMP bands.

In vitro endothelial tube formation assay

To do in vitro tube formation assay, a 24-well plate was coated with ECM Matrigel (Sigma-Aldrich, USA) on ice and incubated at 37°C for 30 min. Then, 3000 human umbilical vein endothelial cells (HUVECs) (National Cell Bank of Iran affiliated to Pasteur Institute of Iran) were cultured in DMEM/F12 containing the 3% FBS and 50-µmol pantoprazole at 37°C, in a humidified incubator for 24 h. The tube-like structure formation was visualized by an inverted microscope (Olympus-IX71-Olympus, USA), and the number of branch points was counted in several fields.

Western blot analysis

Untreated GCSCs and treated GCSCs with 50-µmol pantoprazole cultured at 37°C for 24 h. Untreated and treated GCSCs were centrifuged, and their pellets were treated in lysis buffer composed of 0.5% sodium dodecyl sulfate (SDS), 50-mM Tris base, 1% NP-40, 150-mM NaCl, 2-mM ethylenediaminetetraacetic acid (EDTA) (Merck, Germany), and 1-mM phenyl methane sulfonyl fluoride (PMSF) (Sigma-Aldrich, USA) at 4°C for 5 min. Then, the lysates were centrifuged at 12000 g at 4°C for 15 min. The lysate cells were boiled for 5 min in loading buffer and electrophoresed in 10% SDS-polyacrylamide gel. The protein bands were transported to a polyvinylidene difluoride (PVDF) membrane (Roche, Germany) by transfer buffer that consists of 10-mM Tris, 100-mM glycine, and 10% methanol. PVDF membranes were stained by Ponceau S (Merck, Germany) to confirm the success in the transfer of protein bands. Blots were protected by 5% skimmed milk powder (Merck, Germany) in phosphate-buffered saline-Tween (PBS-T) for 2 h. Immunoblotting was performed by specific primary antibodies and β-actin as internal control overnight and horseradish peroxidase-conjugated secondary antibodies for 2 h. In the end, the protein bands were visualized by the electrochemiluminescence detection kit (Cytomatin, Iran).

Statistical analysis

Results of each test were executed in the triplicate and evaluated in SPSS software (Version 16: SPSS. Link. USA) by Student’s t-test. Student’s t-test was displayed as the mean ± SEM. The P value less than 0.05 was considered as statistically significant.

RESULTS

GCSCs taken from a gastric cancer cell line, MKN-45

To isolate GCSCs, MNK-45 cells were cultivated on a nonsticky surface for 14-16 days. In the early days, tumor cells underwent apoptosis and died, and only cancer stem-like cells proliferated to form the colonies of GCSCs [Figure 1]. The stemness capacity of GCSCs was approved by assessing of stem cell marker expression and self-renewal ability. Data were shown in the previous investigation.[17]

Figure 1.

Isolation of GCSCs from MKN-45 cells by formation of spheroid body in a nonadherence surface. (a) MKN-45 cells cultured on a thin layer of agarose for 3 days. (b) Formation of colonies of GCSCs after 14 days

Pantoprazole reduced cell viability of GCSCs

The influence of pantoprazole on the cell viability of GCSCs was evaluated by the Trypan blue test. GCSCs were given a treatment by various concentrations of pantoprazole (0, 20, 40, 60, 80, and 100 µmol) for 24 h. The results revealed that the cell viability of treated GCSCs was reduced by in a dose-dependent manner of pantoprazole [Figure 2].

Figure 2.

Trypan blue test. GCSCs were treated with various concentrations of pantoprazole (0, 20, 40, 60, 80, and 100 µmol) for 24 h. Results of Trypan blue assay revealed a decrease of cell viability in a dose-dependent manner. (P < 0.05 vs. control group, Student’s t-test was displayed as the mean ± SEM.)

Pantoprazole decreased the MMP activity in GCSCs

To evaluate the effect of pantoprazole on metastasis in GCSCs, the MMP activity was assessed by zymography. The result of zymography exhibited that MMP-2 and MMP-9 activity in the treated GCSCs with 50-µmol pantoprazole decreased in comparison with untreated GCSCs after 24 h [Figure 3].

Figure 3.

Zymography assay. The GCSCs were given a treatment with 50 µmol of pantoprazole for 24 h, and MMP-2 and MMP-9 activities were evaluated by zymography assay. MMP-2 and MMP-9 activities were decreased (line 1) in comparison with untreated GCSCs as control (line 2)

Pantoprazole decreased in vitro tube formation

To examine the effect of pantoprazole on in vitro tube formation, we cultured HUVECs on the ECM Matrigel in 24-well plates and treated them with 50-µmol pantoprazole for 24 h. The results of in vitro tube formation showed that untreated HUVECs differentiated into capillary tubes, but the tube formation was significantly reduced in treated HUVECs with 50-µmol pantoprazole for 24 h [Figure 4].

Figure 4.

In vitro tube formation. The GCSCs were treated by 50 µmol of pantoprazole for 24 h, and the angiogenesis potential of pantoprazole was evaluated by culturing HUVECs on Matrigel with a) 50 µmol of pantoprazole and b) untreated cells

Pantoprazole modified the gene expression of stem cell markers in GCSCs

To evaluate the effect of pantoprazole on stem cell characteristics in GCSCs, we assessed the transcription of stem cell genes including CD44, SOX2, Oct3/4, KLF4, and NANOG by quantification real-time RT-PCR. The RNA level of CD44, Oct3/4, and NANOG in treated GCSCs with 50-µmol pantoprazole for 24 h in comparison with untreated GCSCs was decreased to 0.33, 0.28, and 0.24, respectively. However, the gene expression of SOX2 and KLF4 in treated GCSCs with 50-µmol pantoprazole for 24 h was increased to 4.67 and 3.3 times higher than the untreated GCSCs [Figure 5].

Figure 5.

Quantitative real-time RT-PCR. The GCSCs were given in treatment by 50µmol of pantoprazole for 24 h, and transcription of some genes was measured as follows: The gene expression of CD44, Oct3/4, NANOG, WNT1, and CTNNBP genes at RNA level was declined and the RNA level of SOX2, KLF4, SMARCD1, CTNNBIP1, SUFU, and KREMEN1 genes was increased in pantoprazole treatment of GCSCs. The experiments were performed in three independent triplicate tests. (P value was less than 0.05 vs. the control group, and Student’s t-test was displayed as the mean ± SEM)

Pantoprazole influenced the Wnt signaling pathway

To monitor the influence of pantoprazole on Wnt signaling pathway in GCSCs, we evaluated transcription of genes involved in the Wnt pathway at RNA and protein levels. The transcript level of SMARCD1, CTNNBIP1, SUFU, and KREMEN1 in treated GCSCs with 50-µmol pantoprazole for 24 h was increased to 4.1, 2.3, 1.7, and 3.7 times higher than the untreated GCSCs, but the gene expression of WNT1 and CTNNBP1 in treated GCSCs with 50-µmol pantoprazole for 24 h was decreased in comparison with untreated GCSCs to 0.36 and 0.15, respectively [Figure 5]. We also used Western blot to measure the impact of pantoprazole on the protein level of KREMEN1, SMARCD1, and CTNNBIP1. The results revealed that the protein level of KREMEN1, SMARCD1, and CTNNBIP1 was increased in treated GCSCs with 50-µmol pantoprazole in comparison with untreated GCSCs [Figure 6].

Figure 6.

Western blotting. The GCSCs were given in treatment by 50 µmol of pantoprazole for 24 h, and the Western blot band showed that the protein level of KREMEN1, SMARCD1, and CTNNBIP1 was increased. β-actin was used as the internal control

DISCUSSION

In this study, GCSCs were isolated from GC cell lines by formation of spheroid body in a nonadherence surface that was a very simple and inexpensive strategy.[17] Here, the aim of our study was to examine the impact of pantoprazole on angiogenesis, metastasis, and Wnt signaling pathway in GCSCs taken from MKN-45 cell line. We found that pantoprazole reduced MMP-2 and MMP-9 activities in the GCSCs. We showed that the transcript levels of stem cell markers such as SOX2, Oct3/4, CD44, NANOG, and KLF4 are modified following treatment of GCSCs by pantoprazole. Also, we measured the impact of pantoprazole on transcription and translation of several factors related to Wnt signaling pathway.

Gastric cancer stem cells have a critical role in the tumor initiation, homeostasis, metastasis, and recurrence of gastric cancer.[7] Isolation of cancer stem cells (CSCs) is a critical step to study and reveal the function and mechanisms of CSCs.[7] There are several strategies for isolation and characterization of CSCs, including strategies based on surface markers and fluorescence-activated cell sorting (FACS) and intracellular enzyme activity, which need instruments such as flow cytometry and are also very expensive.[18] Isolation of GCSCs in a nonadherence surface and formation of spheroid body was inexpensive strategy.

Pantoprazole is currently used for treating erosive esophagitis, Zollinger–Ellison syndrome and gastroesophageal reflux, and also eradication of Helicobacter pylori and prevention of peptic ulcer.[19] Drug repurposing or the use of approved drug for new targets and diseases is a growing interest among researchers, especially for curing cancers.[8] Several studies have been performed on the effect of pantoprazole on gastrointestinal cancers, and in some reports, there is some inconsistency in their results. For instance, a study by J. L. Schneider et al. revealed that use of pantoprazole for a long time did not increase the risk of gastric cancer or other gastrointestinal cancers.[20] T-cell–originated protein kinase (TOPK) induces tumorigenesis in several cancers and acts as an oncogene. Treatment of colorectal cancer cells with pantoprazole inhibited the growth of colorectal cancer cells through suppressing of TOPK activity both in vitro and in vivo.[21] Doxorubicin, an anticancer drug, is restricted in acidic organelles of solid tumors and cannot be extended to distal cells, thus causing drug resistance in tumors. Combination use of pantoprazole and doxorubicin in tumor cells showed that pantoprazole increased endosomal pH of tumor cells and, therefore, enhanced the distribution of doxorubicin in solid tumor and improved therapeutic effect of doxorubicin treatment.[22] Cheung indicated that long-term use of pantoprazole for more than three years increased the gastric cancer risk.[23] But, Brunner et al. by following-up studies for 15 years of patients with peptic ulcer or reflux esophagitis did not observe any gastric cancer cases among patients, which treated by pantoprazole.[24,25]

In addition to inconsistent results, a few investigations were performed on the effect of pantoprazole on gastric cancer stem cells. An investigation by Lu et al. performed on the effect of pantoprazole on migration and invasion of the human oral epidermoid carcinoma cell line indicated that pantoprazole had an inhibitory effect on invasion and migration in the human oral epidermoid carcinoma cell line.[14] Also, Lee showed that pantoprazole induced in vitro angiogenesis.[26] But, in another study, Hahm measured the effect of pantoprazole on angiogenesis and expression of angiogenic factors at protein level and showed that pantoprazole significantly reduced in vitro angiogenesis similar to our results of the effect of pantoprazole on the in vitro tube formation in HUVECs.[27]

Transcription of Oct4, CD44, and NANOG, which induced self-renewal capacity and stemness features of GCSCs, was decreased under treatment of pantoprazole in GCSCs. This reduction of cancer stem cell features might be due to the induction of apoptosis or the decline in the cell proliferation of cancer stem cells. The transcript levels of SOX2 and KLF4 were elevated in treated GCSCs by pantoprazole. Carrasco showed that SOX2 protein had a vital function in suppressing gastric tumor growth, and the decline of SOX2 gene expression might induce gastric cancer growth.[28] Also, we investigated the effect of pantoprazole on RNA and protein levels of several factors connected to the Wnt signaling pathway. The Wnt signaling pathway is linked to cell proliferation, apoptosis, migration, and angiogenesis. We previously showed PlGF knockdown and ibuprofen-induced apoptosis and decreased cell proliferation in GCSCs through Wnt signaling pathway.[15,29] The transcription of WNT1 and CTNNB1 decreased in treated GCSCs with pantoprazole in comparison with untreated GCSCs. Our finding on WNT1 and CTNNB1 gene expression approved other studies that confirmed the inducible role of WNT1 and CTNNB1 in gastric cancer progression.[9,30] Moreover, the results of real-time RT-PCR of KREMEN1, SMARCD1, CTNNBIP1, and SUFU genes on the treated GCSCs with pantoprazole indicated an increase in their RNA levels. Also, Western blot results confirmed that protein level of KREMEN1, SMARCD1, and CTNNBIP1 was increased. CTNNBIP1 binds to catenin and inhibits Wnt1 signaling. SMARCD1, KREMEN1, and SUFU act as tumor suppressor genes, and their upregulation reduces cell proliferation and carcinogenesis.[31,32,33,34]

The result of pantoprazole on cell functions such as cell proliferation, apoptosis, metastasis, and angiogenesis and GCSC features indicated that pantoprazole acts as an antitumor drug. However, some studies should be investigated in future. Such as the in vivo study of pantoprazole on gastric cancer for a long period to find probable side effects of drug or investigation on the use of pantoprazole in combination with other anticancer drugs in chemotherapy.

CONCLUSION

In conclusion, finding new therapeutic targets for the existing drugs may be an important strategy for curing diseases. Based on our investigations, pantoprazole may knock down the Wnt1 signaling pathway and reduce tumorigenicity in GCSCs taken from MKN-45 cell line, and pantoprazole may be an excellent therapeutic option for gastric cancer. Nevertheless, further studies have been performed on the influence of pantoprazole on gastric cancer in animals and humans for a long period to uncover probable side effect of pantoprazole treatment.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

This study was supported by the Vice Chancellor of Research Department of Shiraz University of Medical Sciences by code No 17924.