Inhibitory Effect of Astaxanthin on Gene Expression Changes in Helicobacter pylori-Infected Human Gastric Epithelial Cells

Helicobacter pylori (H. pylori) infection promotes gastric carcinogenesis by increasing oxidative stress, inflammation, and dysregulation of cell survival and proliferation of gastric epithelial cells. Astaxanthin (ASTX), a bioactive carotenoid, exhibits antioxidant and anticancer effects by modulating aberrant signaling pathways that lead to dysregulation of cell death and proliferation. To elucidate the molecular mechanism of H. pylori-induced gastric carcinogenesis and to examine the inhibitory effect of ASTX on H. pylori-induced gastric epithelial cell gene expression changes, we performed comparative RNA-sequencing (RNA-Seq) analysis for H. pylori-infected gastric epithelial cells treated with or without ASTX. RNA-Seq results reveal that differentially expressed genes (DEGs) in H. pylori-infected cells were mainly associated with the Wnt/β-catenin signaling pathway, which is related to cell proliferation. ASTX significantly reversed H. pylori-induced transcriptional alterations of the key mediators involved in β-catenin signaling, notably, porcupine (gene symbol, PORCN), spermine oxidase (SMOX), bone morphogenetic protein (BMP) and activin membrane-bound inhibitor (BAMBI), SMAD family member 4 (SMAD4), transforming growth factor-β1 (TGFB1), Fos-like 1 (FOSLI), and c-myc (MYC). We suggest that ASTX may be a potential therapeutic agent that can suppress H. pylori-induced proliferation-associated gene expression changes, in part, by counter-regulating the Wnt/β-catenin signaling pathway.


Introduction
Gastric cancer is the third leading cause of cancer-related mortality worldwide [1]. Helicobacter pylori (H. pylori) infection is a major risk factor for the development of gastric cancer [2]. H. pylori infection can promote gastric carcinogenesis by elevated oxidative stress, inflammation, apoptosis, and autophagy and dysregulation of cell survival and proliferation [3]. H. pylori virulence factors such as cytotoxin-associated protein (cag A) and vacuolization cytotoxin (vac A) play a major role in the pathogenesis of gastroduodenal disease. However, the molecular signaling network leading to H. pylori infection-induced gastric carcinoma is not fully elucidated. To date, several H. pylori-triggered oncogenic signaling pathways have been identified-namely, epidermal growth factor receptor (EGFR), β-catenin, activator protein 1 (AP-1), and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) [4,5]. Increased proliferation of gastric epithelial cells is a hallmark feature in gastric carcinogenesis, and the Wnt/β-catenin signaling pathway is the most prominent axis associated with H. pylori-induced hyper-proliferation [5].
The evolutionarily conserved Wnt/β-catenin signaling controls embryo development and tissue homeostasis. However, aberrant mutations affecting the Wnt/β-catenin signaling pathway often lead to dysregulation of homeostasis and development of cancers. Under basal conditions, cytosolic β-catenin is constantly degraded by the destruction complex AXIN-adenomatous polyposis coli gene (APC)-glycogen synthase β (GSK3β) that targets β-catenin for proteasomal degradation [6]. However, when Wnt ligand is activated, the a cell/bacterium ratio of 1:100 for 4 h based on our previous study on H. pylori-induced gastric proliferation, where mRNA expression of c-myc and cyclin E was significantly increased starting from the 4 h culture time point [18]. Gene expression of H. pylori-infected cells (HP) was compared with that of uninfected AGS cells (None) and H. pylori-infected cells pretreated with ASTX (ASTX + HP). For the ASTX group, AGS cells were treated with ASTX for 7 h.

Preparation of Total RNA Extracts and Library Construction
Total RNA was extracted using the TRI reagent (Molecular Research Center, Cincinnati, OH, USA) and purified and concentrated using the RNeasy MinElute Cleanup Kit (Qiagen, Valencia, CA, USA) according to the manufacturer's protocol. Upon determining the total RNA concentration in each extract, the extracts from three replicate experiments were pooled for RNA-Seq library construction. RNA libraries were constructed and compiled using the QuantSeq 3 mRNA-Seq Library Prep Kit (Lexogen, Vienna, Austria) according to the manufacturer's instructions. The RNA library was newly generated solely for the current experiment using the aforementioned experimental conditions, disparate from our previous RNA-Seq experiment on ASTX in H. pylori-infection [19].

RNA-Sequencing and Bioinformatics Analysis
The total RNA library was subjected to transcriptome sequencing performed by e-Biogen (www.e-biogen.com, Seoul, Korea, accessed on 12 October 2021). QuantSeq 3 mRNA-Seq reads were aligned using Bowtie2. Bowtie2 indices were either generated from the genome assembly sequence or the representative transcript sequences for aligning to the genome and transcriptome. The alignment file was used for assembling transcripts, estimating their abundances, and detecting differential expression of genes. As QuantSeq focuses on the 3 end of polyA RNA, only one fragment per transcript is generated and length normalization is not required. Transcript expression values were quantified based on read counts. Differentially expressed genes (DEGs) were determined based on counts of unique and multiple alignments using coverage in Bedtools. The read count data were processed based on the quantile normalization method using edgeR (http://bioconductor. org, accessed on 12 October 2021). A cutoff was applied at a normalized gene expression (log2) of 2.
Excel-based DEG analysis software (ExDEGA; www.e-biogen.com, Seoul, Korea, accessed on 12 October 2021) was used for data mining and data visualization. Genes with >1.5-fold change in transcript levels were considered to be differentially expressed. Gene classification was based on searches performed using DAVID (http://david.abcc.ncifcrf. gov/, accessed on 12 October 2021) and Medline databases (http://www.ncbi.nlm.nih. gov/, accessed on 12 October 2021). To identify 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, accessed on 12 October 2021).

Validation of DEGs Using Real-Time Polymerase Chain Reaction (PCR)
Real-time PCR was performed to validate the expression profiles of key DEGs identified using RNA-Seq analysis. Candidate genes were selected in relation to the category of molecular pathway. Total RNA was converted to cDNA 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 then 95 • C for 5 min. Real-time PCR assay was subsequently performed using the cDNA templates and primers specific for human. Sequences of the primers used were 5 -CATCCTCATCTACCTACTCAT-3 (forward) and 5 -CGCATCTTGTGCCATGTC-3 (reverse) for porcupine; 5 -CCTACCCTCTCAACGACAGC-3 (forward) and 5 -TAACTACCTTGGGGGCCTTT-3 (reverse) for c-myc; 5 -GACCACAAT CACGACACTGG-3 (forward) and 5 -TTAGCACACCTAGCGACACG-3 (reverse) for spermine oxidase (Smox); 5 -GGCAGCATCACAGTAGCATC-3 (forward) and 5 -GATCGCCAC TCCAGCTACAT-3 (reverse) for bone morphogenetic protein (BMP) and activating membrane-bound inhibitor (Bambi); and 5 -CATCTGAGTCTAATGCTACC-3 (forward) and 5 -CAACAGTCCTTCACTATGG-3 (reverse) for SMAD family member 4 (Smad4). The cDNA was amplified for 40 cycles comprising the following steps: denaturation at 95 • C for 30 s, annealing at 51 • 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. Concurrently, the β-actin gene that served as the reference gene was amplified using the primers 5 -ACCAACTGGGACGACATGGAG-3 (forward) and 5 -GTGAGGATCTTCATGAGGTAGTC-3 (reverse). The relative gene expression of target genes was normalized to that of β-actin.

Statistical Analyses
For the changes in the levels of gene expression in H. pylori-infected AGS cells, all values are expressed as the mean ± standard error of three samples per group (n = 3). Analysis of variance, followed by Neuman-Keuls post hoc test, was used for statistical analyses. Differences were considered statistically significant at p-value < 0.05.

Gene Expression Profile of H. pylori-Infected and/or ASTX-Treated AGS Cells
Gene expression profile of H. pylori-infected AGS cells (HP) was analyzed and compared with that of uninfected AGS cells (None) and H. pylori-infected cells pretreated with ASTX (ASTX + HP). To examine the possible transcriptional effect of ASTX on AGS cells, the gene expression profile of AGS cells treated with ASTX only (ASTX) was also monitored. RNA-Seq analysis on the four experimental groups identified a total of 606 DEGs. The observed DEGs of paired groups are summarized in the Venn diagram and visualized in an MA plot ( Figure 1).

Gene Ontology (GO) Annotation and Functional Analysis of DEGs
To elucidate the biological processes that may be associated with H. pylori stimulation of gastric epithelial cells, DEGs in the first pair (None vs. HP) were aligned according to functional categories using GO annotation. Regulation of transcription (GO:0006355), apoptotic process (GO:0006915), inflammatory response (GO:0006954), and regulation of cell proliferation (GO:0008284, GO:0008285) were the most commonly observed GO annotation terms for biological processes (Figure 2A). pared with that of uninfected AGS cells (None) and H. pylori-infected cells pretreated with ASTX (ASTX + HP). To examine the possible transcriptional effect of ASTX on AGS cells, the gene expression profile of AGS cells treated with ASTX only (ASTX) was also monitored. RNA-Seq analysis on the four experimental groups identified a total of 606 DEGs. To identify the potential mediators of H. pylori infection, overlapping DEGs of HP and None groups were filtered out. H. pylori upregulated 250 genes and downregulated 136 genes in AGS cells. RNA-Seq results revealed that interleukin 8 (gene symbol, CXCL8), chemokine ligand 2 (CXCL1), macrophage inflammatory protein 2 (CXCL2), EGFR, heparin-binding EGF-like growth factor (HBEGF), transforming growth factor-β1 (TGFB1), Smox (SMOX), NF-κB (NFKB1 and NFKB2), TNF-α induced proteins (TNFAIP2 and TNFAIP3), TNF-α induced protein 8 (TNFAIP8), porcupine (PORCN), S-phase kinase-associated protein 2 (SKP2), polycystin2 (PKD2), Kruppel-like factors (KLF5 and KLF6), activating transcription factor 3 (ATF3), Fos-like 1 (FOSL1), and c-myc (MYC) genes were upregulated in AGS cells following H. pylori infection. Since these genes are associated with inflammation and proliferation, it could be inferred that H. pylori-induced inflammation and proliferation may induce carcinogenesis.
To investigate the regulatory effect of ASTX pretreatment on H. pylori stimulation, DEGs from the ASTX + HP group were analyzed, of which 88 genes were upregulated and 68 genes were downregulated. Among the genes differentially regulated by ASTX in H. pylori-infected cells, insulin-like growth factor 1 receptor (IGF1R), gesolin (GSN), his-  Figure 2B).
To discover the inhibitory mechanism of ASTX on H. pylori-induced transcriptional alterations, expression of genes that were counter-regulated by H. pylori and ASTX, either increased by H. pylori infection but decreased in ASTX-pretreated cells or vice versa, were chosen as candidate DEGs of particular interest. Figure 3 illustrates a heatmap representing the expression patterns of the candidate DEGs. Pathway analysis of the candidate DEGs showed that the genes are mostly involved in the CTCF and Wnt signaling pathways (BioCarta), as well as pathways in cancer and Wnt signaling pathways (KEGG).
GO functional annotation was then performed on DEGs in the ASTX + HP group. Regulation of transcription (GO:0006355) and regulation of cell proliferation (GO:0008284, GO:0008285) were markedly enriched biological processes in H. pylori-infected AGS cells pretreated with ASTX ( Figure 2B). To discover the inhibitory mechanism of ASTX on H. pylori-induced transcriptional alterations, expression of genes that were counter-regulated by H. pylori and ASTX, either increased by H. pylori infection but decreased in ASTX-pretreated cells or vice versa, were chosen as candidate DEGs of particular interest. Figure 3 illustrates a heatmap representing the expression patterns of the candidate DEGs. Pathway analysis of the candidate DEGs showed that the genes are mostly involved in the CTCF and Wnt signaling pathways (BioCarta), as well as pathways in cancer and Wnt signaling pathways (KEGG).

Molecular Pathway Analysis of DEGs Involved in Cell Proliferation
Based on the GO and functional analysis, we narrowed our focus to candidate DEGs that were associated with the regulation of cell proliferation via Wnt/β-catenin signaling.  Table 1 shows the list of DEGs reported to be associated with the Wnt/β-catenin pathway, along with the normalized read count of each gene.

Molecular Pathway Analysis of DEGs Involved in Cell Proliferation
Based on the GO and functional analysis, we narrowed our focus to candidate DEGs that were associated with the regulation of cell proliferation via Wnt/β-catenin signaling.  Table 1 shows the list of DEGs reported to be associated with the Wnt/β-catenin pathway, along with the normalized read count of each gene. The PORCN gene, which encodes the Wnt protein activator porcupine, was upregulated in H. pylori-infected cells, but ASTX significantly reduced its expression. Transcriptional expression of Fos-like 1 (FOSL1) and c-myc (MYC) were also significantly increased by H. pylori but suppressed by ASTX. Both Fos-like 1 and c-myc are direct downstream target genes of β-catenin [20]. Our results suggest that the β-catenin pathway was stimulated in our experimental setting and that ASTX treatment may have downregulated the H. pylori-induced β-catenin signaling.
Smox is a polyamine oxidase that generates H 2 O 2 as a byproduct of its catabolic activity. A group of researchers demonstrated that H. pylori increased Smox expression, thereby activating β-catenin signaling in gastric epithelial cells, which is in line with our hypothesis [21]. In this study, we found that SMOX expression was markedly elevated in H. pylori-infected cells and that ASTX treatment significantly decreased the H. pylori-induced SMOX expression.
Using computational modeling based on the Wnt/β-catenin pathway of the KEGG database and relevant research references, a schematic outline of β-catenin signaling in gastric epithelial AGS cells is presented in Figure 4. When Wnt ligands are secreted, which depends on the activity of porcupine, Wnt protein binds to its Frizzled-lipoproteinrelated protein (LRP) receptor. Activated Wnt receptor complex blocks the β-catenin destruction complex (Axin, adenomatous polyposis coli gene (APC), glycogen synthase kinase β (GSK3β), casein kinase 1 (CK1)) from phosphorylating β-catenin, which in turn frees β-catenin from proteasomal degradation by ubiquitin ligase β-Trcp. β-Catenin accumulates and translocates into the nucleus to bind TCF/LEF, and it induces expression of its target genes, such as activating transcription factor 3 (ATF3), c-myc, cyclin D, and Fos-like 1 (FOSL1).
In summary, expression of Wnt protein activators such as porcupine activates the Frizzled/LRP receptor to sequester the destruction complex of Axin, APC, and GSK3β. H. pylori stimulation alters Bambi, PDK2, SKP2, Smox, and Smad4 to fully activate β-catenin signaling. Activated β-catenin signaling promotes cell proliferation via effector genes such as ATF3, FOSL1, and c-myc. Spermine oxidase (Smox) has been indicated in H. pylori-induced gastric carcinogenesis, which may be mediated by β-catenin. SMAD family member 4 (Smad4) and bone morphogenetic protein and activating membrane-bound inhibitor (Bambi), associated with the transforming growth factor-β (TGF-β) signaling pathway, may inhibit β-catenin signaling in H. pylori-infected AGS cells.
In summary, expression of Wnt protein activators such as porcupine activates the Frizzled/LRP receptor to sequester the destruction complex of Axin, APC, and GSK3β. H. pylori stimulation alters Bambi, PDK2, SKP2, Smox, and Smad4 to fully activate β-catenin signaling. Activated β-catenin signaling promotes cell proliferation via effector genes such as ATF3, FOSL1, and c-myc.

PCR Validation of Key Regulatory Genes Involved in H. pylori-Induced β-catenin Pathway
To verify the transcriptional expression patterns of the DEGs, real-time PCR for the representative genes was performed. Figure 5 shows that H. pylori infection increased the mRNA levels of porcupine, c-myc, and Smox, but downregulated Bambi and Smad4 expression. In contrast, ASTX pretreatment protected AGS cells against H. pylori-induced changes in gene expression. However, ASTX treatment alone had little or no effect on the mRNA levels.
To verify the transcriptional expression patterns of the DEGs, real-time PCR for the representative genes was performed. Figure 5 shows that H. pylori infection increased the mRNA levels of porcupine, c-myc, and Smox, but downregulated Bambi and Smad4 expression. In contrast, ASTX pretreatment protected AGS cells against H. pylori-induced changes in gene expression. However, ASTX treatment alone had little or no effect on the mRNA levels.

Discussion
H. pylori infection of the stomach lining promotes gastric carcinogenesis by promoting inflammatory responses that lead to increased cell proliferation and migration. H. pylori secretes virulence factors such as VacA and Cag A to promote chronic inflammation, dysregulation of early immune responses in host cells, and ectopic host cell proliferation

Discussion
H. pylori infection of the stomach lining promotes gastric carcinogenesis by promoting inflammatory responses that lead to increased cell proliferation and migration. H. pylori secretes virulence factors such as VacA and Cag A to promote chronic inflammation, dysregulation of early immune responses in host cells, and ectopic host cell proliferation and apoptosis, as well as disruption of cell junctions and cell polarity [5]. Cellular signaling transduction pathways, such as EGFR, NF-κB, MAPK, PI3K/Akt, and Wnt are the major modulators of H. pylori-induced gastric carcinogenesis [27]. Previously, we observed upregulated activation of the β-catenin pathway, which subsequently led to cell proliferation of H. pylori-infected AGS cells [10,18]. In our preliminary study, we verified the activation of β-catenin, and we presumed that ASTX could inhibit H. pylori-induced β-catenin signaling.
In this study, RNA-Seq analysis was performed to explore the regulatory modulators of H. pylori-induced gastric oncogenesis and the possible effect of ASTX on H. pylori-induced oncogenic malignancies. In particular, GO annotation and molecular pathway analysis identified that DEGs in H. pylori-infected cells were notably associated with the Wnt/βcatenin signaling pathway. Among the various pathways involved in H. pylori-induced oncogenic transformation, Wnt/β-catenin signaling is the most extensively studied pathway that is directly correlated with aberrant epithelial proliferation [28]. ASTX treatment counter-regulated approximately half of the DEGs in H. pylori-infected cells, notably the genes associated with cancer pathways. Previous studies have demonstrated that ASTX effectively inhibits the proliferation of carcinoma cells by suppressing the NF-κB and Wnt/β-catenin pathways [29,30].
RNA-Seq analysis revealed that H. pylori induced overexpression of Fos-like 1, c-myc, TGF-β1, Smox, and porcupine. Specifically, activation of β-catenin led to elevated Foslike 1 and c-myc expression, which collectively promotes cell proliferation and survival. Porcupine, which activates Wnt proteins, leads to elevated proliferation, migration, and invasion of gastric cancer cells via β-catenin signaling activation [31]. Our RNA-Seq results further revealed that ASTX treatment significantly repressed the expression of Fos-like 1, c-myc, and porcupine, which was validated using PCR analysis.
Smox, which catabolizes the polyamine spermine into spermidine and H 2 O 2 , has been extensively studied for its involvement in the development of cancer [32,33]. H. pylori infection elevates Smox expression in gastric epithelial cells, thereby inducing oxidative DNA damage and gastric dysplasia [34,35]. Further study showed that H. pylori-induced β-catenin accumulation was not observed in Smox-/-mice. Administration of Smox inhibitor not only repressed β-catenin activation but also blocked the expression of βcatenin target genes, indicating that Smox promotes H. pylori-induced carcinogenesis via β-catenin signaling. [21] Our results indicate that ASTX significantly inhibited H. pyloriinduced Smox expression.
ASTX treatment reversed the H. pylori-induced downregulation of other genes, including Bambi and Smad4. Bambi was originally known to be a positive regulator of the oncogenic β-catenin pathway [36]. β-catenin stimulates Bambi transcription to inhibit growth-suppressive TGF-β/Smad signaling. However, surprisingly, Bambi expression was decreased in H. pylori-infected AGS cells. A previous study suggested a tumor-suppressive role of Bambi in hepatitis B virus-induced hepatocarcinoma [37]. Induction of Bambi expression suppressed β-catenin activation, TGF-β1 expression, and reduced growth and proliferation of hepatocarcinoma cells and tumor mass formation in vivo [37]. These findings concur with our observations of reduced Bambi expression and increased TGF-β1 expression in H. pylori-infected AGS cells.
Our study showed that ASTX treatment could significantly reverse H. pylori-induced downregulation of Smad4, which functions as a tumor suppressor in gastric epithelial AGS cells [38]. Inhibition of Smad4 enhanced cell proliferation, viability, and migration, which leads to the development of gastric cancer. By contrast, Smad4 expression abrogated cell proliferation and survival via β-catenin inactivation and transcriptional reduction in β-catenin-dependent genes such as cyclin D1, CD44, and c-MET. [39] The current study has limitations due to lack of biological replicates for each treatment condition. Only one library per each group was prepared for this analysis, although it is recommended to have three technical replicates for each group to avoid false positives. Nevertheless, numerous qualitative PCR analyses on the molecular pathway of interest were performed before and after sequencing. The RNA-Seq analysis results from this study could be used as an exploratory analysis in finding transcriptomic patterns to be further investigated.
In conclusion, ASTX treatment is a potential therapeutic agent to suppress H. pyloriinduced activation of the oncogenic Wnt/β-catenin pathway in gastric epithelial cells. We previously performed a similar RNA-Seq experiment on ASTX and H. pylori infection [19]. In the previous study we examined the short-term effect (1 h) of H. pylori infection (at cellular ratio of 50:1), whereas the current study was aimed for the intermediate stage (4 h), leading to cell proliferation observed in long-term infection (at cellular ratio of 100:1). Although the H. pylori infection ratio and exposure time differed, ASTX was administered under the same experimental conditions. Taken together, it can be suggested that ASTX clearly has an inhibitory effect on H. pylori-induced transcriptomic dysregulation in general. The anti-cancer effect of ASTX is not only based on modifications of cellular signaling pathways but also involves its epigenetic modulations as well [40,41]. Further studies are required in the future to elucidate how ASTX affects gene expressions in H. pylori-induced oncogenesis.  Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data generated or analyzed during this study are included within the published article. Expression value of each gene is included in Table 1.