Transcriptome Analysis of the Inhibitory Effect of Astaxanthin on Helicobacter pylori-Induced Gastric Carcinoma Cell Motility

Helicobacter pylori (H. pylori) infection promotes the metastasis of gastric carcinoma cells by modulating signal transduction pathways that regulate cell proliferation, motility, and invasion. Astaxanthin (ASTX), a xanthophyll carotenoid, is known to inhibit cancer cell migration and invasion, however the mechanism of action of ASTX in H. pylori-infected gastric epithelial cells is not well understood. To gain insight into this process, we carried out a comparative RNA sequencing (RNA-Seq) analysis of human gastric cancer AGS (adenocarcinoma gastric) cells as a function of H. pylori infection and ASTX administration. The results were used to identify genes that are differently expressed in response to H. pylori and ASTX. Gene ontology (GO) analysis identified differentially expressed genes (DEGs) to be associated with cell cytoskeleton remodeling, motility, and/or migration. Among the 20 genes identified, those encoding c-MET, PI3KC2, PLCγ1, Cdc42, and ROCK1 were selected for verification by real-time PCR analysis. The verified genes were mapped, using signaling networks contained in the KEGG database, to create a signaling pathway through which ASTX might mitigate the effects of H. pylori-infection. We propose that H. pylori-induced upregulation of the upstream regulator c-MET, and hence, its downstream targets Cdc42 and ROCK1, is suppressed by ASTX. ASTX is also suggested to counteract H. pylori-induced activation of PI3K and PLCγ. In conclusion, ASTX can suppress H. pylori-induced gastric cancer progression by inhibiting cytoskeleton reorganization and reducing cell motility through downregulation of c-MET, EGFR, PI3KC2, PLCγ1, Cdc42, and ROCK1.


Introduction
Globally, gastric cancer is the fifth most common type of cancer and the third leading cause of death from cancer [1]. Helicobacter pylori infection is the major cause of gastric cancer, accounting for more than 60% of the cases [2]. Gastric cancer metastasizes at a high rate, localizing primarily in the liver, peritoneum, lung, and bone [3].
Metastatic cancer occurs as a result of the invasion of the surrounding stroma by primary tumor cells, followed by intravasation of the primary tumor into lymphatic or blood vessels, and finally, by extravasation and colonization of the tumor cells at a distant organ site [4]. Metastasis is initiated when cancer cells acquire specialized behaviors that allow them to breach the basement membrane of the extracellular matrix (ECM). Cancer cells undergo epithelia-mesenchymal transition (EMT), which enables them to lose intercellular junctions and to gain a migratory cell phenotype [5]. Modulation of cell motility is the determinant step in early local invasion. Cell motility requires continuous turnover of the cytoskeleton and modification of cell-cell and cell-substratum adhesions [6].

Gene Expression Profile of H. pylori-Infected and/or ASTX-Treated AGS Cells
The present study was carried out to identify the cell signaling pathway(s) through which ASTX acts on gastric carcinoma AGS cells to mitigate H. pylori-induced activation of cytoskeleton remodeling, cell motility, and cell migration. RNA-sequencing (RNA-Seq) analysis was performed to determine which genes undergo altered expression in AGS cells in response to ASTX administration and H-pylori-infection. Gene ontology (GO) analysis identified differentially expressed genes (DEGs) to be associated with cell cytoskeleton remodeling, motility, and migration. Among these are DEGs encoding c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1, which we confirmed by executing real-time PCR analysis. Herein, we propose a signaling-network through which ASTX might act to counter the effects of H. pylori infection, and thereby protect against H. pylori-induced gastric cancer metastasis.

GO Analysis of DEGs
The DEGs identified above were further analyzed using GO annotation to identify and categorize each gene according to the respective biological process with which they are associated, namely aging, angiogenesis, apoptotic process, cell cycle, cell death, cell differentiation, cell migration, cell proliferation, DNA repair, extracellular matrix, immune response, inflammatory response, cell motility, and cell secretion. The GO categories shared between the DEGs observed for the experimental systems None and HP are inflammatory response, aging, cell migration, and cell motility (Figure 2A). The GO annotations of DEGs that are differentially regulated in the ASTX + HP vs. HP experimental systems include immune response, apoptotic process, cell migration, and DNA repair ( Figure 2B). Genes differentially expressed by ASTX treatment in the absence of H. pylori infection are associated with cell proliferation, cell migration, the apoptotic process, and the cell cycle ( Figure 2C). Among the DEGs that are strongly upregulated by ASTX treatment are genes that encode superoxide dismutase 1, superoxide dismutase 2, TNF receptor-associated factor 7, and serine/threonine kinase 17a (which regulates cell death and apoptosis), protein kinase N2, protein lipase C gamma (PLCγ), protein phosphatase 2, and ROCK1 (which regulates cell migration and cell death).

GO Analysis of DEGs
The DEGs identified above were further analyzed using GO annotation to identify and categorize each gene according to the respective biological process with which they are associated, namely aging, angiogenesis, apoptotic process, cell cycle, cell death, cell differentiation, cell migration, cell proliferation, DNA repair, extracellular matrix, immune response, inflammatory response, cell motility, and cell secretion. The GO categories shared between the DEGs observed for the experimental systems None and HP are inflammatory response, aging, cell migration, and cell motility ( Figure 2A). The GO annotations of DEGs that are differentially regulated in the ASTX + HP vs. HP experimental systems include immune response, apoptotic process, cell migration, and DNA repair ( Figure 2B). Genes differentially expressed by ASTX treatment in the absence of H. pylori infection are associated with cell proliferation, cell migration, the apoptotic process, and the cell cycle ( Figure 2C). Among the DEGs that are strongly upregulated by ASTX treatment are genes that encode superoxide dismutase 1, superoxide dismutase 2, TNF receptor-associated factor 7, and serine/threonine kinase 17a (which regulates cell death and apoptosis), protein kinase N2, protein lipase C gamma (PLCγ), protein phosphatase 2, and ROCK1 (which regulates cell migration and cell death). The overlap in the biological processes identified by the GO annotation analysis represented in Figure 2 centers on cell proliferation, apoptosis, and migration/motility. As the combined cell migration and cell motility categories make up the largest fraction of shared DEG annotations, we carried out further analysis of the DEGs that fall in these two categories. We were particularly interested in identifying the DEGs within this combined category that are upregulated in AGS cells The overlap in the biological processes identified by the GO annotation analysis represented in Figure 2 centers on cell proliferation, apoptosis, and migration/motility. As the combined cell migration and cell motility categories make up the largest fraction of shared DEG annotations, we carried out further analysis of the DEGs that fall in these two categories. We were particularly interested in identifying the DEGs within this combined category that are upregulated in AGS cells in response to H. pylori infection but are not upregulated in AGS cells treated with ASTX prior to H. pylori infection. Accordingly, the normalized (log2) read counts determined for the DEGs of interest are reported in Table 1, and depicted in the heatmap shown in Figure 3. These results were used to construct a signal transduction pathway that governs cell migration/mobility and that is upregulated in AGS cells in response to H. pylori infection but not upregulated or upregulated to a smaller extent in cells treated with ASTX prior to infection. This pathway is presented and discussed in the following section.
Mar. Drugs 2020, x 5 of 13 in response to H. pylori infection but are not upregulated in AGS cells treated with ASTX prior to H. pylori infection. Accordingly, the normalized (log2) read counts determined for the DEGs of interest are reported in Table 1, and depicted in the heatmap shown in Figure 3. These results were used to construct a signal transduction pathway that governs cell migration/mobility and that is upregulated in AGS cells in response to H. pylori infection but not upregulated or upregulated to a smaller extent in cells treated with ASTX prior to infection. This pathway is presented and discussed in the following section.  Table 1. The color gradient from green to red corresponds to increasing gene expression.   Table 1. The color gradient from green to red corresponds to increasing gene expression.

Functional Pathway Analysis of DEGs Altered by H. pylori Infection but Normalized by ASTX Pretreatment
Having identified the DEGs associated with cell motility and migration that are protected by ASTX from upregulation by H. pylori infection, our next step was to map them onto a signal transduction network, which we generated using the KEGG database and input from published studies of cell motility/migration signaling pathways. As illustrated in Figure 4, signaling is initiated by cytokine CXCL-mediated activation of the chemokine receptor CXCR, or by growth factor-mediated activation of a receptor tyrosine kinase. The cytokines CXCL1, CXCL2, and CXCL3 are among the DEGs through which ASTX mitigates H. pylori-induced upregulation (Table 1). In addition, the MET gene, which encodes the receptor tyrosine kinase c-MET (also known as HGF receptor), is also subject to upregulation by H. pylori and partially suppressed by AGS cell pretreatment with ASTX.

Functional Pathway Analysis of DEGs Altered by H. pylori Infection but Normalized by ASTX Pretreatment
Having identified the DEGs associated with cell motility and migration that are protected by ASTX from upregulation by H. pylori infection, our next step was to map them onto a signal transduction network, which we generated using the KEGG database and input from published studies of cell motility/migration signaling pathways. As illustrated in Figure 4, signaling is initiated by cytokine CXCL-mediated activation of the chemokine receptor CXCR, or by growth factormediated activation of a receptor tyrosine kinase. The cytokines CXCL1, CXCL2, and CXCL3 are among the DEGs through which ASTX mitigates H. pylori-induced upregulation (Table 1). In addition, the MET gene, which encodes the receptor tyrosine kinase c-MET (also known as HGF receptor), is also subject to upregulation by H. pylori and partially suppressed by AGS cell pretreatment with ASTX.

Determinations of c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1 mRNA levels in H. pylori-Infected AGS Cells, with and without ASTX Pretreatment
To compare the transcriptional level of the DEGs encoding c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1 in H. pylori-infected AGS cells with and without ASTX pretreatment, real-time PCR was carried out. The results reported in Figure 5 show that H. pylori infection increases the respective mRNA levels in the absence of ASTX pretreatment, whereas H. pylori infection of ASTX-pretreated cells had little or no effect on the mRNA levels.  To compare the transcriptional level of the DEGs encoding c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1 in H. pylori-infected AGS cells with and without ASTX pretreatment, real-time PCR was carried out. The results reported in Figure 5 show that H. pylori infection increases the respective mRNA levels in the absence of ASTX pretreatment, whereas H. pylori infection of ASTX-pretreated cells had little or no effect on the mRNA levels.

Discussion
Cell migration, which is essential to cancer metastasis, is triggered by coordinated modulation of the cytoskeleton and cell-cell and cell-substratum adhesions. A moving cell extends itself at the front by creating protrusions in the intended direction and by making new focal adhesions with the substratum. On the rear ends of the cell, the cell contracts itself and releases its adhesions from the substratum. H. pylori infection triggers the inflammatory response which directly correlates to higher

Discussion
Cell migration, which is essential to cancer metastasis, is triggered by coordinated modulation of the cytoskeleton and cell-cell and cell-substratum adhesions. A moving cell extends itself at the front by creating protrusions in the intended direction and by making new focal adhesions with the substratum. On the rear ends of the cell, the cell contracts itself and releases its adhesions from the substratum. H. pylori infection triggers the inflammatory response which directly correlates to higher rate of gastric cancer progression by stimulating cell proliferation, migration, angiogenesis, and invasion [44][45][46]. Al-Ghoul et al. investigated the association of cagPAI type IV secretion system in cell motility and IL-8 release, and found out that H. pylori that induced high secretion of IL-8 in AGS cells were the ones that are capable of triggering cell motility response in the cells [44]. mRNA expression of IL-8 was notably increased starting from 1 h incubation with H. pylori, and IL-8 was affecting most of the early phase signal pathways in AGS cells infected with H. pylori, especially around 1 h of infection [47,48]. Major signaling pathways leading to migration and invasion, such as PI3K/Akt, PLCγ, and MAPKs such as p38 and ERK, were activated at early time points of H. pylori infection around 30-60 min [47,49,50]. The downstream effectors such as c-Met, Rac1, and Cdc42 were detected in AGS cells at 1 h after H. pylori infection [22,51,52]. In this study, preliminary data showed marked increase of IL-8 mRNA expression level from 1 h incubation with H. pylori, and thus we aimed to find out which signaling networks are turned on during the first phase response of short-term H. pylori infection.
GO analysis of the DEGs identified by RNA-Seq analysis of ASTX-pretreated, H. pylori-infected AGS cells, identified 23 DEGs that are associated with cytoskeleton-remodeling, cell-motility, and cell-migration signaling pathways (Figure 3, Table 1). We focused our attention on five of these, namely the c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1 encoding genes, and demonstrated that their transcription is upregulated by H. pylori, but significantly less so in ASTX-pretreated AGS cells ( Figure 5).
The HGF receptor c-Met is known to be an upstream regulator of cell migration [53]. It is highly expressed in cancer cells, where it promotes actin cytoskeleton rearrangement and cell motility [54]. It has also been demonstrated that micro-RNA-499a-mediated suppression of non-small cell lung cancer cell migration and invasion occurs via downregulation of c-MET gene expression [55]. In addition, Churin et al. reported that expression of the c-MET gene is activated within 30 min following H. pylori infection of AGS cells, and that this activation is required for H. pylori-induced cell motility and scattering [49]. We found that ASTX treatment significantly reduces upregulated c-MET gene expression in H. pylori-infected AGS cells (Table 1, Figure 3). Thus, ASTX might target c-Met to suppress AGS cell migration.
The downstream regulators PI3K and PLCγ facilitate the cell motility cascade by modulating the level of phosphatidylinositol 4,5-bisphosphate (PIP2) [56][57][58]. PI3K generates phosphatidylinositol 3,4,5-trisphosphate (PIP3) from PIP2, or generates PIP2 from phosphatidylinositol 4-phosphate (PIP), thereby facilitating Rho GEF activation and membrane targeting. PI3K is an essential mediator of H. pylori-induced AGS cell migration [59]. In previous studies, ASTX treatment reduced invasive behavior of breast cancer cells and oral carcinoma cells by inhibiting PI3K signaling [42,43]. In a recent study carried out with oral squamous cell carcinoma models, Kowshik et al. reported that ASTX inhibits hallmarks of cancer by targeting the PI3K/NF-κB/STAT3 signaling axis [43]. PLCγ is also an important mediator of cell migration in response to growth factor receptor activation [60,61]. PLCγ modulates the level of PIP2, which in turn governs the activation of Rho family proteins as well as actin-modifying proteins such as the actin-severing cofilin [62]. In the present study, we showed that ASTX suppresses upregulation of the transcription of PI3KC2 and PLCγ1, and their downstream targets ROCK1 and Cdc42 resulting from H. pylori infection of AGS cells.

Cell Line and Culture Conditions
The human gastric adenocarcinoma cell line AGS (ATCC CRL 1739) was purchased from the American Type Culture Collection (ATCC; Rockville, MD, USA). AGS cells were grown in RPMI 1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY, USA), 100 U/mL penicillin, and 100 µg/mL streptomycin. The cells were cultured at 37 • C under a humidified atmosphere consisting of 95% air and 5% CO 2 .

Treatment of AGS Cells with ASTX
ASTX (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO, USA) and stored under nitrogen at −80 • C. Before use, the ASTX stock solution was thawed at 70 • C and added to FBS to achieve the desired concentration. Prior to their infection with H. pylori, AGS cells were preincubated with pure DMSO diluted by FBS at 0.05% final concentration (v/v) (vehicle control) or with 5 µM ASTX for 3 h. The indicated dose and incubation time of ASTX were chosen based on our previous study that demonstrated the inhibitory effect of ASTX on mitochondrial dysfunction and inflammation in H. pylori-infected AGS cells [63]. In particular, 5 µM ASTX exhibited antioxidative effects in AGS cells when preincubated for 3 h.

Bacterial Strain and H. pylori Infection
The H. pylori strain NCTC 11637 was obtained from ATCC. The bacterial cells were grown on chocolate agar plates (Becton Dickinson Microbiology Systems, Cockeysvile, MD, USA) at 37 • C, under microaerophilic conditions, using an anaerobic chamber (BBL Campy Pouch ® System, Becton Dickinson Microbiology Systems, Franklin Lakes, NJ, USA). AGS cells were seeded and cultured overnight to reach 80% confluency. The H. pylori was harvested from the plates, suspended in antibiotic-free RPMI 1640 medium supplemented with 10% fetal bovine serum, and then added to the AGS cell culture at a cellular ratio of 50:1. AGS cells (1.5 × 10 5 /mL) were pretreated with 5 µM ASTX or the control vehicle DMSO for 3 h prior to H. pylori stimulation. The cells were then incubated with H. pylori for 1 h (for preparation of RNA extracts and for determination of mRNA gene expressions).

Preparation of Total RNA Extracts and Library Construction
Total RNA extract was isolated using TRI reagent (Molecular Research Center, Cincinnati, OH, USA) and then cleaned and concentrated using the RNeasy MinElute Cleanup Kit (Quiagen, Valencia, CA, USA) according to the manufacturer's protocol. Following determination of the total RNA concentration in each extract, the extracts from three replicate experiments were pooled for RNA-Seq library construction.
AmpliSeq libraries were constructed and sequenced using the Ion Torrent S5 TM XL next-generation sequencing system (ThermoFisher Scientific, Waltham, MA, USA) in conjunction with the Ion AmpliSeq Transcriptome Human Gene Expression Kit, according to the manufacturer's instructions. For each sample, 30 ng of total RNA was used for cDNA library preparation. Multiple libraries were multiplexed and clonally amplified using the Ion Chef System (Thermo Fisher Scientific, Waltham, MA, USA), and then sequenced. The ampliSeq RNA Plug in (ver 5.6.0.3) by Torrent Suite Software (Thermo Fisher Scientific, Waltham, MA, USA) was used for data analysis.

RNA-Sequencing and Bioinformatics Analysis
The total RNA library was subjected to transcriptome sequencing carried out by e-Biogen (www.e-biogen.com, Seoul, Korea). ExDEGA (Excel-based Differentially Expressed Gene Analysis) software (e-Biogen, Seoul, Korea) was used for initial data processing and analysis for differentially expressed genes (DEGs).
Differential gene expression analysis was performed with ExDEGA v1.6.8 software and using a cutoff at the normalized gene expression (log2) of 2 and p-value of <0.05. DEGs were identified based on >1.5-fold change observed in transcript levels. After filtering the DEGs, Gene Ontology (GO) analysis was performed using the DAVID bioinformatics program (https://david.ncifcrf.gov), for gene identification and annotation. To identify the functional groups and molecular pathways associated with the observed DEGs, RNA-Seq data were further analyzed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (www.genome.jp).

Validation of DEGs by Real-Time Polymerase Chain Reaction (PCR)
Real-time PCR was performed to validate the expression profiles of DEGs identified from RNA-Seq analysis. Candidate genes were selected in relation to the category of function. Total RNA was isolated using TRI reagent (Molecular Research Center, Cincinnati, OH, USA). Conversion of the RNA to cDNA was carried out by incubating the RNA sample with a random nucleotide hexamer and MuLV reverse transcriptase (Promega, Madison, WI, USA) at 23 • C for 10 min, 37 • C for 60 min, and 95 • C for 5 min. The cDNA was used for real-time PCR with primers specific for human. The sequences of the primers used are 5 -TGCACAGTTGGTCCTGCCATGA-3 (forward) and 5 -CAGCCATAGGACCGTATTTCGG-3 (reverse) for c-MET, 5 -TTGTCTGTCACACTTCTGTAGTT-3 (forward) and 5 -AACAGTTCCCATTGGATTCAACA-3 (reverse) for PI3KC2β, 5 -TCGTATATCAGCCAAGGACC-3 (forward) and 5 -AGTACTGGCTTCCAAGAAGG -3 (reverse) for PLCγ1, 5 -GATGGTGCTGTTGGTAAA-3 (forward) and 5 -TAACTCAGCGGTCGTAAT-3 (reverse) for Cdc42, 5 -ATGAGTTTATTCCTACACTCTACCACTTTC-3 (forward) and 5 -TAACATGGCATCTTCGACGACACTCTAG-3 (reverse) for ROCK1. For PCR amplification, the cDNA was amplified by 45 repeat denaturation cycles at 95 • C for 30 s, annealing at 55 • C for 30 s, and extension at 72 • C for 30 s. During the first cycle, the 95 • C step was extended to 3 min.
The β-actin and 18S rRNA gene was amplified in the same reaction to serve as the reference gene.

Statistical Analysis
For the changes in the levels of mRNAs encoding c-MET, PI3KC2β, PLCγ1, Cdc42, and ROCK1 in H. pylori-infected AGS cells, all values are expressed as the mean ± S.E. For each experiment, the number of each group was three (n = 3 per each group). A Student's t-test was used for statistical analysis. A p-value of 0.05 or less was considered statistically significant.

Conclusions
Transcriptional array profiling by RNA-Seq analysis of H. pylori-infected AGS cells, and H. pylori-infected AGS cells pretreated with ASTX was carried out to identify potential mediators in the inhibitory mechanism of ASTX on H. pylori-induced inflammatory and carcinogenic responses. Our findings indicate that ASTX can suppress H. pylori-induced gastric cancer progression by inhibiting cytoskeleton reorganization and reducing cell motility through downregulation of c-MET, EGFR, PI3KC2, PLCγ1, Cdc42, and ROCK1.