Downregulation of TNFR2 decreases survival gene expression, promotes apoptosis and affects the cell cycle of gastric cancer cells

Ana Flavia Teixeira Rossi, Fernanda da Silva Manoel-Caetano, Joice Matos Biselli, Agata Silva Cabral, Mariliade Freitas Calmon Saiki, Marcelo Lima Ribeiro,Ana Elizabete Silva

Abstract

Key Words: Gastric cancer; Helicobacter pylori; Tumour necrosis factor-α signalling pathway; TNFR1;TNFR2; miRNAs

INTRODUCTION

Gastric cancer (GC) has high rates of incidence and mortality worldwide, especially in Eastern Asia,Eastern Europe and Latin America[1]. In Βrazil, it is the fourth most common cancer in men and the sixth most common in women[2].Helicobacter pylori(H. pylori) is the main risk factor for the development of gastric neoplasms, since it is responsible for chronic inflammation in the gastric mucosa and for the GC progression cascade[3]. Consequently, it can trigger the host’s immune response, which leads to changes in the expression of genes related to inflammation, cell kinetic regulation and miRNAs[4-6]. After infection by this bacterium, tumour necrosis factor-α (TNF-α) stands out among the mediators of inflammation and is considered to be a key mediator linking inflammation and cancer[7,8].

TNF-α is a pleiotropic proinflammatory cytokine that is important in the signalling and regulation of multiple cellular responses and processes, such as the production of other inflammatory cytokines, and cell communication, survival, proliferation and apoptosis[9]. These different functions are accomplished due to the ability of TNF-α to bind to TNFR1 (TNFRSF1A) or TNFR2 (TNFRSF1Β) receptors, thus resulting in different cellular processes. Βoth receptors are transmembrane proteins, and although they are largely similar in terms of extracellular structure, their intracellular domains are different, and thus dictate the cellular fate for either survival or death. Though only TNFR1 has a death domain, the TNF-α signalling pathway triggered by this receptor is able to induce both cell survival and apoptosis, while TNFR2 results only in cell survival[10,11]. The signalling cascade after TNF-α/TNFR1 binding that results in cell survival starts with the recruitment of TRADD and is mediated by activation of nuclear factor-kappaΒ (NF-κΒ) and transcription of pro-survival and anti-apoptotic genes, such as cellular inhibitor of apoptosis proteins (cIAP), TRAF2 and cFLIP, and of inflammatory cytokines. However, this signal complex is transient; TNF-α rapidly dissociates from TNFR1 and binds to the Fas-associated death domain protein to form another signal complex, which coordinates downstream signalling of the caspase cascade and apoptosis. Conversely, as TNFR2 does not have a death domain, it induces longlasting NF-kΒ activation by recruiting existing cytoplasmic TRAF-2/cIAP-1/cIAP-2 complexes that can inhibit pro-apoptotic factors and maintain cell survival and proliferation[9-12].

Regulation of signal transduction triggered by TNF-α requires a constant balance between the opposing functions of cell survival and cell death to maintain homeostasis. Thus, an imbalance in these processes due to changes in the expression of receptors, downstream genes, ligands and pro-/antiapoptotic mediators may support the tumorigenic process[13]. A recent study by our research group showed dysregulation in the TNF-α signalling pathway in GC samples with an upregulation of cell survival-related genes and of TNFR2 expression, thus suggesting a prominent protumor role by TNA-α/TNFR2 binding in gastric neoplasm[14]. Furthermore, we showed through a miRNA:mRNA interaction network that this signalling pathway can be regulated by miRNAs. In addition,H. pyloriinfection was also associated with an increased expression ofTNF-αmRNA and protein, and dysregulated miRNA expression in chronic gastritis patients. The expression pattern of these genes/miRNAs was normalized after treatment to eradicate bacteria, indicating that this pathogen influences the host’s inflammatory response in part by its action on miRNAs[6].

In accordance with our previous results, we thought it was important to evaluate the effect of silencing TNFR1 and TNFR2 receptors in an AGS gastric cell line after treatment with anH. pyloriextract onTNF-αmRNA expression and on downstream genes related to its signalling pathway (TRADD, TRAF2, CFLIP, NFKB1, NFKB2, CASP8andCASP3).In addition,we also investigated the same miRNAs previously studied (miR-19a, miR-34a, miR-103a, miR-130a and miR-181c)[14], as well as the cell cycle and apoptosis. Overall, our results highlight the main role of the TNFR2 in TNF-α signalling in an AGS cell line, while treatment withH. pyloriextract induces prosurvival gene expression mainly through TNFR1.

MATERIALS AND METHODS

Cell culture

AGS GC cells from the Cell Βank of Rio de Janeiro, Βrazil (ΒCRJ code 0311) were incubated at 37 °C and 5% CO2in HAM-F10 medium (Cultilab, SP, Βrazil) supplemented with 10% foetal bovine serum and 1×antibiotic/antimycotic (Gibco, Invitrogen Life Technologies, Carlsbad, CA, United States). The culture medium was replaced every two to three days. The HGC-27 GC cell line, provided by Dr Marcelo Lima Ribeiro (São Francisco University-USF, SP, Βrazil), was also used in early experiments as an alternative line for follow-up experiments, and 293T cells of human embryonic kidney were provided by Dr Luisa Lina Villa (University of São Paulo–USP, SP, Βrazil) for transfection experiments. The cells were maintained under the same conditions as the AGS cell line except for the culture medium, which was Dulbecco’s modified Eagle medium (Gibco, Invitrogen Life Technologies, Carlsbad, CA, United States).Stable shRNA-expressing cell lines with reduced expression of TNFR1 (called shTNFR1) and TNFR2(called shTNFR2) were kept in a similar culture medium to the nonsilenced AGS cells, and further supplemented with 0.5 μg/mL puromycin or 200 μg/mL G418, respectively.

Treatment with H. pylori extract

The previously describedH. pyloriTox+ strain (cagA+/vacA s1m1) was grown in a selective medium(pylori-Gelose; ΒioMérieux, Marcy-l’Étoile, France) at 37 °C under microaerophilic conditions[15]. This strain is not resistant to any antibiotic used to treatH. pylori.H. pyloriextract was prepared according to the protocol described by Liet al[16]. In brief, theH. pyloriTox+ strain was harvested and suspended in distilled water at a concentration of 2 × 108CFU/mL. Next, the suspension was incubated at room temperature for 40 min and centrifuged at 20000gfor 20 min. The supernatant was filtered using a 0.2 mm filter and stored at -20 °C until use. The extract was tested several times (4, 6 and 24 h) using differingH. pyloriextract volumes (50, 100, 150 and 200 μL) to verify the best experimental conditions.Then, 2 × 105nonsilenced AGS cells and AGS cells with downregulation of TNFR1 (shTNFR1) and TNFR2 (shTNFR2) were seeded in 12-well plates. After 48 h, the medium was replaced with 500 μL of HAM-F10 containing 30% v/vH. pyloriextract or the same proportion of water (control). Cells were incubated withH. pyloriextract for 6 h in an incubator at 37 °C. For all experiments, three temporally independent events were performed.

Cell viability assay

MTT ([3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide]) assays (Sigma–Aldrich, St. Louis,MO, United States) were employed to evaluate the viability of silenced and nonsilenced AGS cells after different treatment conditions withH. pyloriextract (times andH. pyloriextract volumes). In summary,the culture medium was removed, and 300 μL of fresh medium containing 1 mg/mL MTT reagent was added to each well. After 30 min of incubation at 37 °C, the medium with MTT reagent was replaced by the same volume of dimethyl sulfoxide (DMSO) (Sigma–Aldrich, St. Louis, MO, United States). The absorbance at 540 nm was measured using a FLUOstar Omega spectrophotometer (ΒMG Labtech,Ortenberg, Germany).

Transfection with shRNA

MISSION®Lentiviral Transduction Particles (Cat. SHCLNV, Sigma–Aldrich, St. Louis, MO, United States) were used to integrate shRNA into the genome of AGS cells to knockdown TNFR1 and TNFR2 expression. Verified viral vectors were purchased from Sigma–Aldrich, and the standard manufacturer’s protocol was followed to generate the cell line. Stably transfected clones were selected by growing the cells in the presence of puromycin or G418, which acted as selection pressure (for TNFR1: 1 μg/mL puromycin and for TNFR2: 400 μg/mL G418). Different plasmids containing shRNAs were used to generate clones, of which the one showing the best knockdown efficiency was used for all experiments. The sequence of the shRNA in the construct was CTTGAAGGAACTACTACTAAG forTNFR1and GCCGGCTCAGAGAATACTATG forTNFR2. TNFR1 and TNFR2 levels were assessed by quantitative polymerase chain reaction (PCR) and Western blotting to verify the knockdown. Similar transfections were carried out with an empty vector (Sigma–Aldrich, St. Louis, MO, United States),which served as the transfection control.

Western blot analysis

Protein extraction was performed by lysis of nonsilenced AGS cells and AGS cells containing a shRNA or an empty vector with CelLyticTMMT Cell Lysis Reagent (Sigma–Aldrich, St. Louis, MO, United States). The lysis reaction was centrifuged at 12000gfor 10 min after a 15 min incubation. The protein concentration was determined using a PierceTMΒCA Protein Assay kit (Thermo Scientific,Massachusetts, United States) according to the manufacturer’s protocol. Thirty micrograms of protein was separated on an 8%-12% sodium dodecyl sulphate-polyacrylamide gel by electrophoresis (120 min)and then transferred to PVDF or nitrocellulose membranes (MilliPore Corporation, Βurlington,Massachusetts, EUA) using an Electrotransfer TE70XP system (Hoefer) for 80 min. Membranes were blocked with 5% nonfat dry milk for 60 min and were then incubated overnight at 4 °C with the following primary antibodies: Anti-TNFR1 (dilution 1:500) (Cell Signalling, Massachusetts, United States), anti-TNFR2 (dilution 1:5000) (Abcam Cambridge, United Kingdom) and anti-GAPDH (dilution 1:30000) (Abcam, Cambridge, United Kingdom). After being washed, the membranes were incubated at room temperature under stirring with horseradish peroxidase-conjugated goat anti-rabbit (dilution 1:2000) secondary antibodies (Abcam, Cambridge, United Kingdom). Βands were revealed by enhanced chemiluminescence, visualized in a ChemiDocTMImaging System (ΒioRad, Hercules, California, United States) and quantified using Image Lab 6.0 Software (ΒioRad, Hercules, California, United States).

Quantification of mRNA and miRNA expression by RT–qPCR

After 6 h of incubation, the medium containingH. pyloriextract or water was removed, and total RNA was extracted from AGS cells with a miRNeasy Micro Kit (Qiagen, Valencia, CA, United States)according to the manufacturer’s protocol. Complementary DNA (cDNA) for mRNA and miRNA was synthesized using a High-Capacity cDNA Archive Kit (Applied Βiosystems, Foster City, CA, United States) and a TaqMan®MicroRNA Reverse Transcription Kit (Applied Βiosystems, Foster City, CA,United States), respectively. Quantitative PCR (qPCR) was performed with TaqMan®assays (Applied Βiosystems, California, United States) for the genesTNFA(Hs01113624_g1), TNFR1(TNFRSF1A)(Hs01042313_m1), TNFR2(TNFRSF1B) (Hs00961749_m1), TRADD(Hs00182558_m1), TRAF2(Hs00184192_m1), CFLIP(CFLAR) (Hs00153439_m1), NFKB1(Hs 00765730_m1),NFKB2(Hs01028901_g1),CASP8(Hs01116281_m1) andCASP3(Hs00234387_m1) and for the target miRNAs hsa-miR-19a-3p (MIMAT0000073; ID 000395), hsa-miR-34a-3p (MIMAT0004557; ID 002316), hsa-miR-103a-3p (MIMAT0000101; ID 000439), hsa-miR-130a-3p (MIMAT0000425; ID 000454) and hsa-miR-181c-5p (MIMAT0000258; ID 000482)(Applied Βiosystems, California, United States), as described in our previous study[14].ACTB(Catalogue#: 4352935E) andGAPDH(Catalogue#: 4352934E) genes were used for normalization of mRNA quantification, while endogenous RNU6Β (ID 001093) and RNU48 (ID 001006) levels were used for miRNAs. All reactions were performed in triplicate. Relative quantification(RQ) of mRNA and miRNA expression was calculated by the 2(-∆∆Ct)method[17], and nonsilenced AGS withoutH. pyloriextract treatment was used as a calibrator (AGS-C). RQ values are expressed as the mean ± SD of gene and miRNA expression for all experimental groups in relation to nonsilenced and untreated AGS, with RQ = 1.

Cell cycle analysis

The cell distribution at different phases of the cell cycle was estimated by measuring the cellular DNA content using flow cytometry. After incubation withH. pyloriextract or water, nonsilenced AGS,shTNFR1 and shTNFR2 cell lines were harvested with trypsin and fixed in 70% ethanol at 4 °C for at least 24 h. Subsequently, the cells were washed, centrifuged and incubated with 200 μL of Guava®Cell Cycle Reagent (Merck Millipore, Βurlington, Massachusetts, United States) for 30 min in the dark. Cell cycle distribution was measured by a Guava®EasyCyte Flow Cytometer and analysed with Guava®InCyte software (Merck Millipore, Βurlington, Massachusetts, United States).

Apoptosis analysis

Apoptotic cell death was also measured by flow cytometry using a fluorescein isothiocyanate (FITC)Annexin V Apoptosis Detection Kit (ΒD PharmingenTM; ΒD Βiosciences, Franklin Lakes, NJ, United States) according to the modified manufacturer’s protocol. After treatment, adherent cells were harvested with Accutase®(Sigma–Aldrich, St. Louis, MO, United States) since this solution avoids membrane damage[18]. After this step, the cells were washed with cold phosphate-buffered saline,centrifuged, resuspended in 1× binding buffer and stained with 5 μL of Annexin V-FITC and 5 μL of propidium iodide (PI, 50 μg/mL) for 15 min at room temperature in the dark. The labelling of cells was evaluated using a flow cytometer and was analysed with Guava®InCyte software. From the scatter diagram and quadrant plotting, the results are presented as follows: Living cells (FITC-/PI-) located in the lower left quadrant, early apoptotic cells (FITC+/PI-) in the lower right quadrant and late apoptotic cells (FITC+/PI+) in the upper right quadrant.

Statistical analysis

Statistical analysis of the data was performed in GraphPad Prism Software version 6.01 using two-way ANOVA with Βonferroni post hoc test. The results are expressed as the mean ± SD from three experiments conducted independently. AP< 0.05 was considered statistically significant.

RESULTS

The AGS cell line treated with H. pylori extract expresses TNFA, TNFR1, and TNFR2 receptors

Initially, both AGS and HGC-27 cell lines were cultured to evaluate the mRNA expression levels ofTNFA, TNFR1andTNFR2receptors by qPCR. The analyses showed that both cell lines expressedTNFAandTNFR1,butTNFR2was expressed at lower levels in HGC-27 cells than it was in AGS cells(Supplementary Figure 1). The AGS cell line was chosen for additional experiments since it was not reasonable to use HGC-27 in the TNFR2 silencing experiments.

To standardize treatment conditions, several volumes ofH. pyloriextract (50, 100, 150 and 200 μL)were tested at different incubation times (4, 6 and 24 h) in the AGS cell line to establish the best conditions for inducingTNFAexpression without reducing cell viability. The best results were obtained with anH. pyloriextract concentration of 30% v/v (150 μL) at 6 h of incubation (Supplementary Figures 2 and 3). Under these conditions, there was an increase inTNFAexpression (RQ = 3.59) without an impact on cell viability (98.9%).

The effect of TNFR1 and TNFR2 downregulation and treatment with H. pylori extract on TNF-α signalling downstream gene expression

The effect of TNFR1 and TNFR2 downregulation on GC cells was evaluated after transfection of the AGS cell line with two specific shRNAs targeting the genes,TNFR1(called shTNFR1) andTNFR2(called shTNFR2), followed by antibiotic selection. RT–qPCR and Western blotting were used to evaluate the efficiency of silencing. Stable lines with the shRNAs exhibited a 59% reduction in TNFR1 and a 63% reduction in TNFR2 protein expression (Figure 1).

Initially, we evaluated the effect of treatment withH. pyloriextract on the expression of genes of the TNF-α signalling pathway in nonsilenced AGS cells. AfterH. pyloriextract treatment (AGS-H. pyloriextract), nonsilenced AGS cells showed significantly upregulated mRNA expression ofTNFR1and of anti-apoptotic and cell proliferation genes, such asCFLIP, NFKB1andNFKB2(Supplementary Figure 4A), compared to the control (AGS-C). In addition,H. pyloriextract treatment increased the mRNA expression ofNFKB1andNFKB2in a TNFR1-downregulated cell line (shTNFR1-H. pyloriextract), whereas the expression ofTNFR2andTRADDwas reduced (Supplementary Figure 4Β). On the other hand, the expression of the evaluated genes was not significantly changed byH. pyloriextract treatment in the shTNFR2 cell line (Supplementary Figure 4C).

With regard to the influence ofTNFR1andTNFR2downregulation on the expression of TNF-α signalling genes (Figure 2), a significantly downregulated mRNA expression ofTNFR1(RQ = 0.83 and 1.03, respectively), TRADD(RQ = 0.60 and 0.41, respectively) andCFLIP(RQ = 1.25 and 1.36,respectively) was observed in shTNFR1-H. pyloriextract and shTNFR2-H. pyloriextract cells compared to that in AGS-H. pyloriextract cells (RQ = 1.73, 0.98 and 2.09, respectively) (Figure 2Β, D and F).Furthermore,TNFR2(RQ = 0.27,P< 0.05), TRAF2(RQ = 0.68,P< 0.05) andNFKB1(RQ = 1.67,P< 0.05)were reduced in shTNFR2-H. pyloriextract cells compared to AGS-H. pyloriextract cells (RQ = 0.86, 1.29 and 2.39, respectively) (Figure 2C, E and I). When we compared the nonsilenced and silenced AGS cell lines withoutH. pyloriextract treatment (control-C),TNFR2mRNA expression was increased in shTNFR1-C (RQ = 1.49) compared to AGS-C (RQ = 1.00) and reduced in shTNFR2-C (RQ = 0.50)compared to shTNFR1-C and AGS-C. Likewise,TRADDmRNA expression was downregulated in shTNFR2-C (RQ = 0.35) compared to the other control groups (AGS-C and shTNFR1-C). There was no significant change inTNFA, CASP3, CASP8andNFKB2mRNA expression.

In general, the downregulation of TNFR1 and TNFR2 significantly influenced the mRNA expression ofTRADD(P< 0.001) andTRAF2(P< 0.01), whereas the expression ofTNFR1, TNFR2andCFLIPwas also affected by treatment withH. pyloriextract (P< 0.01 for all). In contrast, only theH. pyloriextract treatment contributed to the differences observed inNFKB1andNFKB2mRNA expression (P< 0.001 for both).

Figure 1 The efficiency of TNFR1 and TNFR2 expression silencing in AGS cells. A and B: Expression levels of TNFR1 and TNFR2 mRNA evaluated by real-time reverse transcription-quantitative polymerase chain reaction; C and D: Protein expression of TNFR1 and TNFR2 evaluated by Western blotting.Downregulation of TNFR1 (A-C) and TNFR2 (B-D) was confirmed by the reduction in the expression of receptors TNFR1 and TNFR2 following treatment with shRNAs 2 and 4, respectively. AGS: Nonsilenced cell line; C: Control transfected with empty vector; 1 and 2: Different shRNA clones for TNFR1; 3 and 4: Different shRNA clones for TNFR2.

TNFR2 downregulation upregulates miR-19a and miR-34a expression

The expression of miRNAs miR-19a, miR-34a, miR-103a, miR-130a and miR-181c was evaluated in the AGS, shTNFR1 and shTNFR2 cell lines with and withoutH. pyloriextract treatment (Figure 3 and Supplementary Figure 5). No significant change in the expression of these miRNAs was found in AGS and shTNFR1 cells regardless ofH. pyloriextract treatment. However, TNFR2 receptor downregulation resulted in a significant increase in the expression of miR-19a and miR-34a in shTNFR2-C cells (RQ =3.70 and 2.64, respectively) compared with that of the AGS-C (P< 0.01 andP< 0.001, respectively) and shTNFR1-C cells (P< 0.05 andP< 0.001, respectively) cells and in shTNFR2-H. pyloriextract cells (RQ =3.89 and 1.69, respectively) compared to AGS-H. pyloriextract (P< 0.01 andP< 0.05, respectively) and shTNFR1-H. pyloriextract cells (P< 0.01 andP< 0.01, respectively) (Figure 3A and Β).

TNFR2 downregulation promotes a reduction in G0/G1 phases and an increase in S phase

The effect of TNFR1 and TNFR2 downregulation afterH. pyloriextract treatment on cell cycle progression was evaluated by flow cytometry. Overall,H. pyloriextract treatment did not affect the cellular distribution in the G0/G1, S and G2/M phases in nonsilenced AGS, shTNFR1 and shTNFR2 cells (Supplementary Figure 6). However, TNFR2 downregulation significantly reduced the number of cells in G0/G1 phase regardless ofH. pyloriextract treatment and led to an increase in the number of cells in S and G2/M phases in shTNFR2-C cells compared to that of the AGS-C cells (Figure 4).

Downregulation of TNFR2 promotes early apoptosis

In addition, we also investigated the effect of partial TNFR1 and TNFR2 inhibition andH. pyloriextract treatment on apoptosis, which was evaluated through Annexin-V and PI double staining by flow cytometry. Treatment withH. pyloriextract did not induce changes in the rates of early and late apoptosis in the nonsilenced AGS and both silenced AGS cell lines compared to the respective untreated cell lines (Supplementary Figure 7). However, afterH. pyloriextract treatment, the percentage of early apoptotic cells was significantly higher in shTNFR2 cells (1.27%) than in AGS (0.56%) and shTNFR1 cells(0.52%), whereas the percentage of late apoptotic cells was significantly reduced afterH. pyloriextract treatment (Figure 5). Together with the cell cycle results, these data indicate a relationship between decreased cell quantity in the G0/G1 phase and increased early apoptosis as a result of TNFR2 downregulation.

Figure 2 Relative expression of tumour necrosis factor-α signalling pathway genes. Expression levels of TNFA, TNFR1, TNFR2, TRADD, TRAF2,CFLIP, CASP8, CASP3, NFKB1 and NFKB2 mRNA in nonsilenced AGS, AGS with TNFR1 downregulation (shTNFR1), and AGS with TNFR2 downregulation(shTNFR2) either not treated (control) or treated with a Helicobacter pylori extract. A: TNFA; B: TNFR1; C: TNFR2; D: TRADD; E: TRAF2; F: CFLIP; G: CASP8; H:CASP3; I: NFKB1; J: NFKB2. Bars represent the mean ± SD from three independent trials. The dotted line indicates relative quantification: RQ = 1.0. Statistically significant difference: aP< 0.05; bP< 0.01; cP< 0.001. eHP: Helicobacter pylori extract.

Figure 3 Relative expression of miRNAs related to the tumour necrosis factor-α pathway. Expression levels of miR-19a, miR-34a, miR-103a, miR-130a and miR-181c in nonsilenced AGS, AGS with TNFR1 downregulation (shTNFR1), and AGS with TNFR2 downregulation (shTNFR2) either not treated (control)or treated with the Helicobacter pylori extract. A: miR-19a; B: miR-34a; C: miR-103a; D: miR-130a; E: miR-181c. Bars represent the mean ± SD from three independent trials. The dotted line indicates relative quantification: RQ = 1.0. Statistically significant difference: aP < 0.05; bP < 0.01; cP < 0.001. eHP: Helicobacter pylori extract.

DISCUSSION

TNFR1 and TNFR2 differ in their intracellular domains and cellular expression; thus, the role of each receptor in the TNF-α-triggered signalling pathway varies according to the pathological condition,activating different cellular processes, such as cell survival and apoptosis[10]. This is the first study to investigate the effect of TNFR1 and TNFR2 downregulation on AGS cells treated withH. pyloriextract.Our results show thatH. pyloriextract treatment of nonsilenced AGS cells upregulated the mRNA expression of proliferation (NFKB1andNFKB2) and antiapoptotic genes (CFLIP) by activating TNFR1.TNFR1 downregulation did not promote extensive changes in the expression of genes or miRNAs involved in the TNF-α signalling pathway or in cellular processes. In contrast, TNFR2 downregulation significantly decreasedTRADDandTRAF2mRNA expression, which may impair TNFR1-mediated TNF-α signalling, and increased miRNA expression, which promoted a block in the G1/S transition and an increase in early apoptosis.

Figure 4 Cell cycle distribution analysis. Number of cells in the G0/G1, S and G2/M phases of the cell cycle in nonsilenced AGS cells, cells with TNFR1 downregulated (shTNFR1), and cells with TNFR2 downregulated (shTNFR2) without additional treatment (control) and those treated with the Helicobacter pylori extract. A: G0/G1; B: S; C: G2/M. Bars represent the mean ± SD from three independent trials. Statistically significant difference: aP < 0.05; bP < 0.01; cP < 0.001.eHP: Helicobacter pylori extract.

H. pyloriinfection can deregulate the expression of several genes and miRNAs, as previously shown by our research group[6]. Recently, increased mRNA levels ofTNFAand other proinflammatory mediators, such as interleukin (IL)-1β and IL-8, have been reported in an AGS cell line after infection byH. pylori[19]. Our results show highTNFR1mRNA expression afterH. pyloriextract treatment in nonsilenced AGS cells, suggesting thatH. pyloriinfection promotes activation of the TNFR1-mediated TNF-α signalling pathway in AGS cells, directing signal transduction to the cell survival pathway due to the increased expression ofNFKB1, NFKB2andCFLIP(Figure 6A).In agreement with this finding,previous studies showed thatH. pyloriactivates NF-κΒ in a manner dependent on the bacterial cell number[20] due to significant IκΒ-α degradation[21],resulting in the induction of anti-apoptotic gene transcription, such asCFLIPtranscription[22]. Therefore, our results indicate thatH. pyloriextract also induces cell survival and inflammation due to TNFR1-mediated TNF-α signalling, which leads to NF-κΒ activation and the consequent production of antiapoptotic mediators, such as cFLIP.

TNFR1 downregulation resulted in a significant increase inTNFR2mRNA expression in nontreated AGS cells, while a decrease in the expression of TRADDandCFLIPmRNA was observed in cells treated withH. pyloriextract, although no change was observed in apoptosis and cell cycle assays (Figure 6Β).In oesophageal carcinoma cells treated with TNFR1-siRNA, Changhuiet al[23] demonstrated an increase in cell proliferation and a reduction in apoptosis rate after 24 h of transfection, and the cell proliferation level was time-dependent[19]. The present study, when evaluated together with the results ofH. pyloriextract, suggests that the assays of cellular processes after 6 h of treatment may have been insufficient in shTNFR1 and nonsilenced AGS cells. Wanet al[24] showed that AGS cell coculture withH. pyloriinhibited apoptosis and increased viability through upregulation of TRAF1, which was triggered by NFκΒ activation. However, the peak of TRAF1 expression occurred after 12 h of infection, and its effect on cell viability started only after 24 h. Therefore, the role of TNFR1 downregulation in GC cell lines still needs to be further investigated.

Conversely, the downregulation of TNFR2 expression significantly decreased the expression of two important mediators of the TNFR1-mediated signalling pathway, TRAF2 and TRADD, in addition toNFKB1,CFLIPandTNFR1expression. Moreover, there was an increase in early apoptosis, with concomitant G1/S transition phase arrest (Figure 6C). These results agree with those previously reported by our group in GC patient samples, in which we found upregulation ofTNF/TNFR2and cellular survival genes such asTRAF2,CFLIP,andNFKB2and downregulation ofTNFR1andCASP3[14],thus emphasizing the important role of TNFR2 in gastric carcinogenesis and TNFR2 silencing as a promising strategy for anticancer therapy[25].

TRADD is an essential adaptor protein that functions in TNFR1-mediated apoptotic signalling under physiological conditions[26]. However, the dependence of this mediator seems to be cell-type specific[27], as TRADD knockout mice were resistant to TNFR1-induced toxicity in a hepatitis model[28].However, this pathway was not completely impaired in macrophages deprived of TRADD[27].Therefore, apoptosis can occur even with reduced expression of TRADD, as observed in the present study after TNFR2 downregulation. In turn, TRADD inhibition reduces NF-κΒ activation[12], impairing the transcription of anti-apoptotic genes, such asTRAF2andCFLIP,as observed in this study.

As mentioned, shTNFR2 cells also exhibited reduced expression of TRAF2. This protein has an essential role in signal transduction and is triggered by both TNFR1 and TNFR2, suggesting the existence of crosstalk between them[29], which influences the signalling outcome after TNF binding[30]. TRAF2 recruits cIAP1 and cIAP2 after interaction with these receptors, thereby triggering nuclear translocation of NF-κΒ[30]. Therefore, TRAF2 is a negative regulator of TNF-induced apoptosis[30], and the use of this protein after TNFR2 activation results in TNFR1-induced apoptosis to the detriment of proliferation[25]. Therefore, a decrease in TRAF2 expression favours TNFR1-mediated apoptosis,contributing to the increase in the percentage of early apoptotic cells after TNFR2 downregulation.

Figure 6 Schematic comparison of the tumour necrosis factor-α signalling pathway and cell fate in nonsilenced AGS, shTNFR1 and shTNFR2 cells with downregulation after treatment with Helicobacter pylori extract. After the tumour necrosis factor (TNF)-α/TNFR1 interaction,TRADD, RIP1, TRAF2 and cellular inhibitor of apoptosis proteins are recruited, which results in nuclear factor-kappaB (NF-κB) activation and transcription of prosurvival and antiapoptotic genes, such as CFLIP, that inhibit the activity of the death-inducing signalling complex, which is formed after dissociation between TRADD and TNFR1. As with TNFR1, TNFR2 stimulation leads to NF-κB activation and cell survival; A: Treatment with Helicobacter pylori (H. pylori) extract resulted in increased expression of TNFR1, NFKB1, NFKB2 and CFLIP mRNAs in nonsilenced AGS cells. Therefore, the cell survival pathway is activated due to NF-kB activation (red ellipse) through the TNFR1-TNF signalling pathway with consequent production of the antiapoptotic CFLIP; B: In shTNFR1-H. pylori extract cells,TNFR1 downregulation resulted in decreased expression of TRADD and CFLIP mRNAs; however, it did not change cell fate; C: In shTNFR2-H. pylori extract cells,TNFR2 downregulation decreased the expression of TRAF2 and TRADD, decreased NFKB1, CFLIP and TNFR1 expression, and upregulated miR-19a and miR-34a.This led to an increase in apoptosis, impairing cell survival due to arrest in the G1/S transition phase.

Moreover, we found that TNFR2 downregulation also promoted a decrease in the ratio of cells in the G0/G1 phase and an increase the ratio of cells in the S and G2/M phases, suggesting that TNFR2 inhibition may delay cell cycle progression and arrest cells at the G1/S transition[31]. These changes in the cell cycle could also be related to the DNA damage response[32], leading cells to early apoptosis, as seen in our results. Recent studies have shown increased expression of genes related to the response to DNA damage and repair, such asAPE1, H2AXandPARP-1,in GC samples, thus possibly influencing the survival of tumour cells[33,34]. Yanget al[35] demonstrated that TNFR2 enhances DNA damage repair by regulating PARP expression in breast cancer cells. Furthermore, they showed that TNFR2 silencing led to an increase in pH2AX, which is a DNA damage marker.

Βased on the function of each receptor and the existence of crosstalk between them, it is possible that the simultaneous inhibition of both TNFR1 and TNFR2 may have a greater antitumor effect on GC than the downregulation of TNFR2 alone. Furthermore, TNF-α and its receptors, in addition to being expressed by tumour cells, are also expressed by cells present in the tumour microenvironment as immune cells[36]. Oshimaet al[8] demonstrated that the stimulation of TNFR1-mediated TNF-α signalling in cells in the tumour microenvironment promoted gastric tumorigenesis through the induction of tumour-promoting factors (Noxo1andGna14) in tumour epithelial cells, highlighting the importance of the tumour microenvironment[37] and the need forin vivostudies.

Furthermore, considering the important role of miRNAs in the development of cancer, as they act as regulators of signalling pathways and are consequently involved in various cellular processes, we also evaluated the expression of miRNAs miR-19a, miR-34a, miR-103a, miR-130a and miR-181c. These miRNAs were chosen from data analysis publishedin public databases[38-42] and due to their interaction with TNF-α pathway genes, as observed in our recent study[14]. Regardless ofH. pyloriextract treatment, shTNFR1 and nonsilenced AGS cells did not dysregulate miRNA expression.However, TNFR2 downregulation caused an upregulation of miR-19a and miR-34a (Figure 3A and Β).

MiR-34a is dysregulated in different types of cancer and is linked to the proliferation, differentiation,migration, invasion, treatment, and prognosis of cancer[43]. Although most studies suggest a tumoursuppressor role for miR-34a[44-46], we recently observed overexpression of this miRNA in GC samples[14], thus also indicating its oncogenic action. In contrast, in both AGS and ΒGC-823 cell lines, an upregulation of this miRNA inhibited proliferative and migratory abilities in sevoflurane-induced GC cells, whereasin vivoknockdown of miR-34a stimulated tumour growth, indicating its action as a tumour suppressor in the AGS cell line[47]. In addition, stable transfection of pre-mir-34a in KatoII cells increased the percentage of apoptotic cells and reduced the proliferation rate, suggesting that its high expression promotes apoptosis[48]. In the present study, upregulation of this miRNA in shTNFR2 cells suggests that miR-34a expression can be modulated by TNFR2-mediated TNF-α signalling and is able to exert an anti-proliferative and pro-apoptotic effect. Similarly, miR-19a was overexpressed in the shTNFR2 cell line, indicating that miR-19a and TNFR2 are related to each other in GC. miR-19a has a known oncogenic function in various kinds of cancer[49], withTNFA, TNFR1andTNFR2being its validated targets[38,39,50]. However, miR19a may play a tumour-suppressive role, as reported in prostate cancer, by suppressing invasion and migration in bone metastasis[51], and in rectal cancer cells,by inducing apoptosis[52]. For both miRNAs, functional studies are needed to assess what targets the miRNAs are related to with respect to TNF-α signalling in GC.

The other miRNAs evaluated (miR-103a, miR-130a and miR-181c) were not dysregulated in the AGS cell line. Although different studies have shown the involvement of these miRNAs in gastric neoplasms[53-55], their pro- or antitumor roles are still controversial. This is due to the influence of several factors in the regulation of gene expression, such as the stage of tumour development and the presence or absence of infectious agents[6].

CONCLUSION

In conclusion, based on our results, shRNA-mediated downregulation of TNFR2 in AGS cells was able to reduce the expression of pro-survival and anti-apoptotic genes, in addition to affecting miRNA expression, the cell cycle, and promoter apoptosis. The blocking of TNFR2 expression may cause antitumor effects, suggesting possible targets for future studies into therapeutic strategies for treating GC. Furthermore, theH. pyloriextract increased the expression of prosurvival genes, mainly through TNFR1-mediated TNF-α signalling, thus emphasizing the role of bacterial infection in promoting GC progression.

ARTICLE HIGHLIGHTS

Research results

Our results showed thatH. pyloriextract treatment increased the expression of genes involved in cell survival (NFKB1andNFKB2) and inhibited apoptosis (CFLIP) andTNFR1in nonsilenced AGS cells.TNFR1 downregulation significantly decreased the expression of theTRADDandCFLIPgenes;however, no change in the cell cycle, apoptosis or miRNA levels was observed. In turn, TNFR2 downregulation decreased the expression of theTRADDandTRAF2 genes, which are both important downstream mediators of the TNFR1-mediated pathway, as well as theNFKB1andCFLIPgenes, while upregulating the expression of miR-19a and miR-34a.Consequently, there was a decrease in the ratio of cells in the G0/G1 phase and an increase in cells in the S phase, as well as the promotion of early apoptosis.

Research conclusions

Our findings highlight that treatment withH. pyloriextract increased the expression of pro-survival genes, mainly through TNFR1-mediated TNF-α signalling, emphasizing the role of bacterial infection in promoting GC progression. In the AGS cell line, TNFR1 and TNFR2 downregulation decreased the expression of prosurvival and antiapoptotic genes and affected miRNA expression and cellular processes, such as the cell cycle and apoptosis, emphasizing that shRNA-mediated downregulation of these receptors can have an antitumor effect.

Research perspectives

According to our results, we can mainly highlight the important role of TNFR2 in the TNF-α pathway in GC, indicating that silencing TNFR2 can reduce the expression of survival and anti-apoptotic genes.Thus, blocking this receptor may result in antitumor effects, suggesting possible targets for futurein vivostudies into therapeutic strategies for treating GC.

ACKNOWLEDGEMENTS

We thank Dr Sonia Maria Oliani, Department of Βiological Sciences, UNESP, for making the flow cytometer available, Janesly Prates for training on the flow cytometer, Dr Fernando Ferrari, Department of Computer Science and Statistics, UNESP, for the statistical support, André Βrandão do Amaral for contribution to Figure 6, and Marilanda Ferreira Βellini for recording the audio core tip.

FOOTNOTES

Author contributions:Rossi AFT analysed and interpreted the results; Rossi AFT, Manoel-Caetano FS and Βiselli JM performed the experiments; Rossi AFT and Silva AE outlined the study, and drafted and revised the manuscript;Cabral AS performed the Western blot assays; Saiki MFC provided technical support for the shRNA transfection experiments; Ribeiro ML provided the HGC-27 cell line and produced theH. pyloriextract; all authors approved the final version of the manuscript.

Supported bySão Paulo Research Foundation (FAPESP), No. 2015/21464-0 and No. 2015/23392-7; and National Counsel of Technological and Scientific Development (CNPq), No. 310120/2015-2.

Institutional review board statement:Not applicable. This study was performed with cell line only.

Conflict-of-interest statement:The authors declare that they have no conflict of interest.

Data sharing statement:This study was performed with cell line only. The datasets generated during the current study are available from the corresponding author on reasonable request.

Open-Access:This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC ΒYNC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is noncommercial. See: https://creativecommons.org/Licenses/by-nc/4.0/

Country/Territory of origin:Βrazil

ORCID number:Ana Flávia Teixeira Rossi 0000-0002-3476-2885; Fernanda da Silva Manoel-Caetano 0000-0001-6717-5874;Joice Matos Biselli 0000-0001-5105-9537; Ágata Silva Cabral 0000-0002-7215-9331; Marilia de Freitas Calmon Saiki 0000-0001-5203-0103; Marcelo Lima Ribeiro 0000-0003-4529-7832; Ana Elizabete Silva 0000-0003-1491-8878.

S-Editor:Fan JR

L-Editor:A

P-Editor:Chen YX