Icariin inhibits gastric cancer cell growth by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 signaling axis

Abstract

This study aimed to clarify the expression and role of hsa_circ_0003159 in gastric carcinogenesis, and validate the protective effects of Icariin (ICA) against gastric cancer (GC) cell growth through the in vitro and in vivo experiments. The levels of hsa_circ_0003159, microRNA (miR)-223-3p and NLRP3 were measured by Quantitative real time Polymerase Chain Reaction or western blot. The cell counting kit (CCK)-8 was used to determine cell proliferation. The target relationship of miR-223-3p/hsa_circ_0003159 and miR-223-3p/NLRP3 was predicted by bioinformatics and validated by the dual-luciferase reporter and pull-down assays. Xenograft model was constructed to assess the roles of hsa_circ_0003159 and protective effects of ICA in GC in vivo. Results showed that hsa_circ_0003159 was downregulated in GC cell lines and its overexpression promoted GC cell viability. MiR-223-3p was identified as a target of hsa_circ_0003159. By competitively sponging miR-223-3p, hsa_circ_0003159 positively regulated NLRP3 expression. MiR-223-3p mimics reversed the suppressive effect of hsa_circ_0003159 on GC cell viability and cell pyroptosis. Importantly, ICA inhibited GC cell viability and triggered GC cell pyroptosis by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 axis in vitro and in vivo. In conclusion, this study indicated ICA inhibits GC cell growth by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 signaling axis. This study not only reveals the mechanism of gastric carcinogenesis but also provides potential molecular targets and therapeutic tools for its treatment.

Introduction

Gastric cancer (GC) is the fifth most diagnosed malignant tumor worldwide, and the number of new cases in China is as high as 400,000 per year, accounting for 40% of the total number of cases in the world. ¹ The 5-year survival rate of early GC can reach 95%, ² but due to the low rate of early diagnosis, most patients are already at an advanced stage during diagnosis, thus missing the best time for surgery, and the prognosis is poor, with a 5-year survival rate of only about 15%.

Icariin (ICA) is the main active ingredient of Epimedium and belongs to the group of flavonoids. ³ The pharmacological activity of ICA has been studied in relation to its anti-inflammatory effects, ⁴ protection of the cardiovascular system, ⁵ anti-osteoporosis,^6,7 neuroprotective, immunoprotective effect and reproductive function. ⁸ In addition, numerous studies have reported that ICA has pharmacological effects such as inhibiting the proliferation and metastasis of various tumor cells. For example, ICA could effectively trigger apoptosis, enhance anti-tumor immunity and inhibit the migration of breast cancer cells, which might serve as a potential candidate drug for the treatment of breast cancer. ⁹ ICA could also potentiate the efficacy of radiotherapy in the murine model of colorectal cancer. ¹⁰ ICA also had the potential to be an adjunct to chemotherapy for osteosarcoma. ¹¹ It seems that ICA exerts protective effects on various tumors via inducing apoptosis, reversing immune escape, enhancing the sensitivity of chemotherapeutic drugs, reducing the toxic side effects of radiotherapy. In GC, ICA increases the proportion of GC cells in G0/G1 phase, which could provide a new strategy for the therapy of GC. ¹² ICA also exerts negative effects on GC cell invasion and migration and maybe a potential anti-cancer drug. ¹³ However, more molecular mechanisms and pathways should be further investigated.

CircRNAs are a class of non-coding RNAs that do not have a 5^′ terminal cap and a 3^′ terminal Poly (A) tail and are characterized by a covalent closed-loop structure.^14,15 Hsa_circ_0003159, which is produced by the CACNA2D1 gene located on chromosome 781689743-81746489, is significantly down-regulated in GC tissues. ¹⁶ Further studies found that hsa_circ_0003159 suppressed GC progression by regulating the miR-223-3p/NDRG1 axis. ¹⁷ In addition, Ren et al. ¹⁸ reported that LncRNA ADAMTS9-AS2 acts as a tumor suppressor and activates NLRP3-mediated GC cell death through secretion of miR-223-3p, thereby enhancing the sensitivity of GC cells to cisplatin. It is thus hypothesized that hsa_circ_0003159 may inhibit GC progression by regulating the miR-223-3p/NLRP3 axis.

In addition, studies have shown that ICA alleviates rheumatoid arthritis by regulating the miR-223-3p/NLRP3 signaling axis. ¹⁹ Taken together, this study was aimed to explore whether ICA may promote GC progression by regulating hsa_circ_0003159/miR-223-3p/NLRP3 signaling axis.

Materials and methods

Reagents and instruments

Human gastric mucosal epithelial cells (GES-1) and GC cell lines (SGC7901, BGC-823, MKN-7, HGC-27, NUGC-3, AGS, HS-746T and N87) were from BeNa Culture Collection, Beijing, China. Roswell Park Memorial Institute (RPMI)-1640, fetal bovine serum (FBS) was from TermoFisher (Wilmington, DE, USA). The penicillin and streptomycin were from Beyotime (Shanghai, China). The vectors for hsa_circ_0003159 overexpression were obtained from YRBIO (Changsha, China). The miR-223-3p mimics and mimic negative control were designed and synthesized by Sangon Biotech (Shanghai, China). Lipofectamine 3000 reagents was purchased from Invitrogen, CA, USA. Cell counting kit-8 (CCK-8) was purchased from AbMole, USA. The Gemini EM microplate reader was purchased from Molecular Devices, USA. The biotin-labeled probes for hsa_circ_0003159 were designed and synthesized from Sangon Biotech (Shanghai, China). The streptavidin Dynabeads, Trizol reagent (15596-018), SuperScript II RNase H reverse transcriptase was from Invitrogen (Carlsbad, CA, USA). The psiCHECK-2 luciferase reporter vector and dual-luciferase assay system were from Promega (Madison, WI, USA). BALB/c nude mice were from Charles River (Beijing, China). RNase R was from Geneseed Biotech (Guangzhou, China). SYBR mixture (RR420A) was from Takara (Dalian, China). The protein assay kit was from Bio-Rad (Hercules, CA, USA). Primary antibodies: β-actin, NLRP3, cleaved caspase-1 and IL-1β were from Abcam (Cambridge, MA, USA).

Cell culture and transfection

Cells were cultured in RPMI-1640 containing FBS (10%) and penicillinm, streptomycin solution (1%) at 37°C in 5% CO₂. Hsa_circ_0003159 overexpression, miR-223-3p mimics (miR-223-3p, 5’-UGUCAGUUUGUCAAAUACCCCA-3’) and mimic negative control (miR-NC, 5’-GUGGAUUUUCCUCUAUGAUUU-3’) were delivered into GC cells by using the Lipofectamine 3000 reagent. 24 h later, different concentrations (0, 10, 20, 40 and 80 μmol/L) of ICA were administrated into BGC-823 and HGC-27 cell lines for different times (0, 12, 24 and 48 h). 48 h later, GC cells were collected for the following analysis.

CCK-8 assay

The GC cell proliferative abilities were measured using the commercial CCK-8 kit. In brief, cells were incubated with CCK-8 reaction solution (4 h) and the Gemini EM microplate reader was used to measure optical density (OD) values (450nm) to evaluate cell proliferation.

Pull-down assay

Cells were fixed, lysed, sonicated and centrifuged. After that, the supernatants were used as input, and the rest part of the BGC-823 and HGC-27 cells were incubated (30°C, overnight) with the streptavidin Dynabeads mixture containing the biotin-labeled probes for hsa_circ_0003159. After that, to reverse crosslinking and release miR-223-3p, the lysis buffer and Proteinase K were employed and finally were quantified by Quantitative real time Polymerase Chain Reaction (QRT-PCR).

Dual-luciferase reporter assay

To predict the potential targets of hsa_circ_0003159 and miR-223-3p, Circular RNA Interactome and StarBase were used respectively. The hsa_circ_0003159 sequences or NLRP3 3’ untranslated region (UTR) containing miR-223-3p binding sites were amplified and cloned into the downstream of psiCHECK-2 luciferase reporter vector to form the wild-type (WT) luciferase reporter vectors. By mutating the binding sites of miR-223-3p, the mutant (MUT) luciferase reporter constructs were generated. For dual-luciferase reporter assay, BGC-823 and HGC-27 cells were co-transfected with these constructs and miR-223-3p or miR-NC. After 48 h, by a dual-luciferase assay system, luciferase activity was measured.

Quantitative real time Polymerase Chain Reaction (QRT-PCR)

Total RNA was extracted from GC cells and tissues using Trizol reagent. RNA was treated by RNase R (20 min) for purity of circRNAs, which was then reverse-transcribed by SuperScript II RNase H reverse transcriptase. 18S rRNA and U6 were regarded as endogenous controls. Specific primers for each gene were as follows: hsa_circ_0003159: Forward, 5’-CCGAACATCTGTCTCCGAAA-3’; Reverse, 5’-CTGCTGCGTGCTGATAAGAT-3’; miR-223-3p: Forward, 5’-AGCTGGTGTTGTGAATCAGGCCG-3’; Reverse, 5’-TGGTGTCGTGGAGTCG-3’; NLRP3: Forward, 5’-CACTTCCAGTTTTTGCCGGG-3’, Reverse, 5’- GGGAATGGCTGGTGCTCAAT -3’; 18S rRNA (Forward, 5’-GGCCCTGTAATTGGAATGAGTC-3’; Reverse, 5’-CCAAGATCCAACTACGAGCTT-3’); U6 (Forward, 5’-CTCGCTTCGGCAGCACATATACT-3’; Reverse, 5’-ACGCTTCACGAATTTGCGTGTC-3’).Synthesized cDNA was mixed with SYBR mixture and subjected to real-time PCR quantification using the ABI PRISM 7700. All reactions were performed in triplicate. The relative expression was calculated by the 2^−ΔΔCt method.

Western blot

Western blot analysis was performed on hepatic tissue and cells, which were homogenized in ice-cold lysis buffer, centrifuged (14,000 rpm, 4°C, 15 min), and the supernatant was collected. Protein concentrations in the supernatants were measured using a protein assay kit. Then, proteins (30g) were subjected to SDS-PAGE electrophoresis, and then transferred to PVDF membranes and blocked in BSA (1 h). After washing in TBST, the membranes were incubated (4°C, overnight) with primary antibodies: β-actin (1:200, ab16039), NLRP3 (1:1000, ab263899), cleaved caspase-1 (1:200, ab138483) and IL-1β (1:1000, ab9722), and incubated with HRP conjugated second antibody (37°C, 1 h). The bands were analyzed with an ECL kit.

Xenograft model

Animal research was conducted under the experimental animal use guidelines and with the approval of the ethics committee of the First Affiliated Hospital of Anhui Medical University. BALB/c nude mice (5-week-old; male) to establish xenograft model as described previously. ¹⁷ In brief, the lentiviral vector of hsa_circ_0003159 was constructed and infected into HGC-27 cells and selected by puromycin to choose the stably transfected cells. HGC-27 cells (2×10⁶ cells per mouse) were then subcutaneously injected into the mice. 7 days after inoculation, when the tumor volume reached approximately 100 mm³, the mice in hsa_circ_0003159 and Vector groups were then randomized equally into 2 subgroups (phosphate buffer solution (PBS) and 40 mg/kg ICA). Mice were treated with ICA intraperitoneally every 2 days for 2 weeks. The volume was measured every 7 days. At the end of the experiment, the mice were euthanized by cervical dislocation. The tumor was isolated, imaged and weighed

Statistics

GraphPad 6.0 software was used for statistical analysis. Data were represented as mean ± SEM. The student’s t-test was used for the comparisons between two groups and one-way analysis of variance (ANOVA) was used for the comparisons among three or more than three groups. p < 0.05 was considered as significant.

Results

Hsa_circ_0003159 was downregulated in GC cell lines and its overexpression promoted GC cell viability

To gain a clear understanding of the functional role of hsa_circ_0003159 in the regulation of GC progression, its expression in GC cell lines (SGC7901, BGC-823, MKN-7, HGC-27, NUGC-3, AGS, HS-746T and N87) was detected by qRT-PCR. Interestingly, the expression level of hsa_circ_0003159 was downregulated in GC cell lines compared to gastric epithelial cell line GES-1 (Figure 1(a)). Among them, BGC-823 and HGC-27 cell lines showed the lowest expression of hsa_circ_0003159, which were used in the following analysis. BGC-823 and HGC-27 cell lines were transfected with overexpression vector carrying hsa_circ_0003159, which was verified by qRT-PCR assay (Figure 1(b)). CCK8 assay showed that hsa_circ_0003159 overexpression prompted the viability of BGC-823 and HGC-27 cell lines (Figures 1(c) and (d)).

Figure 1.

Hsa_circ_0003159 was downregulated in GC cell lines and its overexpression inhibited GC cell viability. (a) Hsa_circ_0003159 expression in GC cell lines (SGC7901, BGC-823, MKN-7, HGC-27, NUGC-3, AGS, HS-746T and N87) was detected by qRT-PCR. (b) BGC-823 and HGC-27 cell lines were transfected with overexpression vector carrying hsa_circ_0003159, which was verified by qRT-PCR assay. CCK8 assay showed that hsa_circ_0003159 overexpression inhibited the viability of BGC-823 (c) and HGC-27 (d) cell lines. *p < 0.5 compared to GES-1 or vector group. GC: gastric cancer. QRT-PCR: Quantitative real time Polymerase Chain Reaction.

Hsa_circ_0003159 directly targeted miR-223-3p

QRT-PCR assay indicated that hsa_circ_0003159 overexpression significantly decreased the expression level of miR-223-3p in BGC-823 and HGC-27 cell lines (Figure 2(a)). Next, to detect the ability of hsa_circ_0003159 to sponge miR-223-3p, pull-down assay showed results, which were validated by the hsa_circ_0003159 probe (Figure 2(b)). Circular RNA Interactome predicted that miR-223-3p had the binding sites of hsa_circ_0003159 (Figure 2(c)). To validate this relationship, the dual-luciferase reporter assay was performed in BGC-823 and HGC-27 cell lines. As shown in Figure 2(d) and Figure 2(e), in the WT-hsa_circ_0003159 group, the luciferase activity was declined more than 60% in BGC-823 and HGC-27 cell lines by miR-223-3p mimics as compared to NC mimics group.

Figure 2.

Hsa_circ_0003159 directly targeted miR-223-3p. (a) QRT-PCR assay indicated that hsa_circ_0003159 overexpression significantly decreased the expression level of miR-223-3p in BGC-823and HGC-27cell lines. (b) Pull-down assay showed that the amount of miR-223-3p pulled down using probe-hsa_circ_0003159 was detected by qRT-PCR and significantly higher than probe-control. (c) Circular RNA Interactome predicted that miR-223-3p had the binding sites of hsa_circ_0003159. The luciferase activity was declined more than 60% by miR-223-3p mimics in wild-type hsa_circ_0003159 group in BGC-823 (d) and HGC-27 (e) cell lines, while it showed little effect in mutant hsa_circ_0003159 group. *p < 0.5 or **p < 0.05 compared to miRNA NC group/Vector group/Probe-control group. QRT-PCR: Quantitative real time Polymerase Chain Reaction.

Hsa_circ_0003159 enhanced NLRP3 expression by regulating miR-223-3p in GC cell lines

In BGC-823 and HGC-27 cell lines, qRT-PCR assay indicated that hsa_circ_0003159 overexpression significantly increased the mRNA (Figure 3(a)) and protein (Figure 3(b)) expression level of NLRP3 as compared to the Vector group. The starBase analysis showed that NLRP3 had the binding sites with miR-223-3p (Figure 3(c)). Dual-luciferase reporter assay showed that miR-223-3p mimics caused a great reduction of luciferase activity in the NLRP3 3’UTR-WT group as compared to NC mimics group (Figures 3(d) and (e)).

Figure 3.

Hsa_circ_0003159 enhanced NLRP3 expression by regulating miR-223-3p in gastric cancer cell lines. In BGC-823and HGC-27cell lines, qRT-PCR and western blotting assays indicated that hsa_circ_0003159 overexpression significantly increased the mRNA (a) and protein (b) expression level of NLRP3 as compared to Vector group. (c) The targets of miR-223-3p were explored by SstarBase, which showed that NLRP3 had the miR-223-3p binding sites.(d-e) Dual-luciferase reporter assay showed that miR-223-3p mimics led to great loss of luciferase activity in NLRP3 3’UTR-WT group, whereas it did not affect the activity in NLRP3 3’UTR-MUT group. *p<0.5 compared to miRNA NC group/Vector group. QRT-PCR: Quantitative real time Polymerase Chain Reaction.

Hsa_circ_0003159 inhibited GC cell proliferation by regulating the miR-223-3p/NLRP3 axis

In BGC-823 and HGC-27 cell lines, qRT-PCR and western blotting assays indicated that hsa_circ_0003159 overexpression significantly increased the mRNA and protein expression levels of NLRP3 as compared to the Control group. On the contrary, miR-223-3p mimics significantly decreased the mRNA and protein expression levels of NLRP3 as compared to the Control group, while co-transfection of hsa_circ_0003159 overexpression and miR-223-3p mimics alleviated the effects of hsa_circ_0003159 overexpression (Figures 4(a)–(c)). CCK8 assay showed that hsa_circ_0003159 overexpression inhibited the viability of BGC-823 and HGC-27 cell lines, whereas miR-223-3p mimics significantly enhanced the viability of BGC-823 and HGC-27 cell lines. Interestingly, co-transfection of hsa_circ_0003159 overexpression and miR-223-3p mimics alleviated the effects of hsa_circ_0003159 overexpression on the viability of BGC-823 and HGC-27 cell lines (Figures 4(d)–(e)).

Figure 4.

Hsa_circ_0003159 inhibited gastric cancer cell proliferation by regulating the miR-223-3p/NLRP3 axis. (a) qRT-PCR and (b–c) western blotting assays indicated that miR-223-3p mimics significantly decreased the mRNA and protein expression levels of NLRP3 as compared to Control group, while co-transfection of hsa_circ_0003159 overexpression and miR-223-3p mimics alleviated the effects of hsa_circ_0003159 overexpression. (d–e) CCK8 assay showed that hsa_circ_0003159 overexpression inhibited the viability of BGC-823 and HGC-27 cell lines, whereas miR-223-3p mimics significantly enhanced the viability of BGC-823 and HGC-27 cell lines. *p < 0.5 compared to Control group.^#p < 0.5 compared to hsa_circ_0003159 group. QRT-PCR: Quantitative real time Polymerase Chain Reaction.

ICA treatment up-regulated hsa_circ_0003159 expression in GC cell lines

Different concentrations (0, 10, 20, 40, 80 μmol/L) of ICA were administrated into BGC-823 and HGC-27 cell lines for different times (0, 12, 24, 48 h) and the cell viability was detected by CCK8 assay. The results showed that the cell survival rate was inhibited by ICA in a time-dependent and given concentration-dependent manner (Figures 5(a) and (b)). ICA caused the upregulation of hsa_circ_0003159, and downregulation of miR-223-3p (Figures 5(c) and (d)). Besides, ICA led to the increased mRNA and protein expression levels of NLRP3 as compared to the Vehicle group (Figures 5(e)–(g)).

Figure 5.

ICA treatment up-regulated hsa_circ_0003159 expressions in GC cell lines. CCK8 assay showed that the survival rate of BGC-823 (a)and HGC-27(b) cell lines was inhibited by ICA in a time-dependent and given concentration-dependent manners. ICA caused the upregulation of hsa_circ_0003159 (c), and downregulation of miR-223-3p (d). ICA led to the increased mRNA (e) and protein expression levels of NLRP3 (f–g) as compared to Vehicle group. *p<0.5 compared to Vehicle group. ICA: Icariin

ICA inhibited GC cell viability and triggered cell pyroptosis by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 axis

CCK8 assay showed that co-treatment of ICA and hsa_circ_0003159 overexpression caused the lowest viabilities in BGC-823 and HGC-27 cell lines (Figures 6(a) and (b)). The western blotting assay showed that co-treatment of ICA and hsa_circ_0003159 overexpression caused the highest protein expression levels of NLRP3, cleaved caspase-1 and IL-1β in BGC-823 and HGC-27 cell lines (Figures 6(c)–(e)). These results indicated that ICA aggravated the effects of hsa_circ_0003159 overexpression on cell viability and protein expression levels of NLRP3, cleaved caspase-1 and IL-1β.

Figure 6.

ICA inhibited gastric cancer cell growth by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 axis. CCK8 assay showed that co-treatment of ICA and hsa_circ_0003159 overexpression caused the lowest viabilities in BGC-823 (A)and HGC-27 (B)cell lines.(C-E)Western blotting assay showed that co-treatment of ICA and hsa_circ_0003159 overexpression caused the highest protein expression levels of NLRP3, cleaved caspase-1and IL-1β in BGC-823and HGC-27cell lines.*p < 0.5 compared to Vehicle +vector group. ^#p<0.5 compared to Vehicle +hsa_circ_0003159 group. ICA: Icariin.

ICA inhibited GC tumor growth by regulating the hsa_circ_0003159 signaling axis

GC cells were infected with hsa_circ_0003159 overexpression vector, screened with puromycin, and then injected subcutaneously into nude mice. One week after inoculation, when the tumor volume reached approximately 100 mm³, mice were treated with 40 mg/kg of ICA intraperitoneally every 2 days for 2 weeks. After that, results showed that tumor volume and tumor weight was the lowest in the group that co-treatment with ICA and hsa_circ_0003159 overexpression (Figures 7(a)–(c)). Besides, consistent with the in vitro results, co-treatment of ICA and hsa_circ_0003159 overexpression caused the highest protein expression levels of NLRP3, cleaved caspase-1 and IL-1β in mice (Figures 7(d)–(e)).

Figure 7.

ICA inhibited gastric cancer tumor growth by regulating the hsa_circ_0003159 signaling axis. (a–c) GC cells were infected with hsa_circ_0003159 overexpression vector, screened with puromycin, and then injected subcutaneously into nude mice. One week after inoculation, when the tumor volume reached approximately 100 mm³, mice were treated with 40 mg/kg of ICA intraperitoneally every 2 days for 2 weeks. After that, results showed that tumor volume and tumor weight were imaged and calculated. (d–e) The protein expression levels of NLRP3, cleaved caspase-1and IL-1β in nude mice with hsa_circ_0003159 overexpression and ICA administration. *p < 0.5 compared to Vector + PBS group. ^#p < 0.5 compared to hsa_circ_0003159 + PBS group. ICA: Icariin, PBS: phosphate buffer solution.

Discussion

GC is a public problem with high morbidity and mortality and circRNAs may play importance in clinical value. Previous work has shown that the downregulation of hsa_circ_0003159 has potential clinical importance in GC, ¹⁶ and is related to lower disease-free survival and overall survival. ¹⁷ However, how and whether hsa_circ_0003159 mediates the progression and development of GC remains unclear. Here, hsa_circ_0003159 was significantly downregulated in GC cell lines and its overexpression promoted GC cell viability. Consistently, a previous study ¹⁷ found that hsa_circ_0003159 inhibited GC cell growth by suppressing cell proliferation, migration and invasion, whereas enhancing GC cell apoptosis, indicating its role act as a tumor suppressor in GC.

Moreover, miR-223-3p was identified as a target of hsa_circ_0003159 and miR-223-3p mimics abated the suppressive effect of hsa_circ_0003159 on GC cell growth. ¹⁷ Here, the target association of miR-223-3p-hsa_circ_0003159 was predicted by bioinformatics and validated by the dual-luciferase reporter and pull-down assays. Co-transfection of hsa_circ_0003159 overexpression and miR-223-3p mimics significantly alleviated the effects of hsa_circ_0003159 overexpression on the viability of BGC-823 and HGC-27 cell lines. This studies indicated that miR-223-3p is a promotor in GC progression, which was consistent with previous observations, showing its pro-proliferative, pro-invasive and anti-apoptotic effects in GC.^20,21

Our further study validated that hsa_circ_0003159 promoted NLRP3 expression by competitively sponging miR-223-3p. The target association of miR-223-3p-NLRP3 was predicted by bioinformatics and validated by a dual-luciferase reporter assay. It has been reported that miR-223-3p mimics could alleviate cell pyroptosis in age-related macular degeneration ²² and endothelial cells ²³ by inactivation of the NLRP3 inflammasome. Besides, in GC, ¹⁸ miR-223-3p-NLRP3 target association is also involved in GC development and sensitizes chemo resistant GC cells to cisplatin.

Moreover, the CCK-8 assay showed that the cell survival rate was significantly inhibited by ICA in a time-dependent and given concentration-dependent manner, showing its protective effects on GC. ICA caused the notable upregulation of hsa_circ_0003159 and NLRP3, and notable downregulation of miR-223-3p. Besides, ICA aggravated the suppressive effects of hsa_circ_0003159 overexpression on GC cell viability, and protein expression levels of pyroptosis associated proteins (NLRP3, cleaved caspase-1 and IL-1β) in vitro and in vivo. It is well-known that factors including viruses, bacteria, fungi, crystal particles and components of dying cells, can activate the NLRP3 inflammasome, which finally results in caspase-1-dependent maturation of pro-IL-1β. And then, pro-IL-1β ultimately induces a cell death mechanism called pyroptosis. ²⁴ This was the first evidence concluding that ICA inhibits GC cell growth, especially GC cell pyroptosis, at least partially, through regulating the hsa_circ_0003159/miR-223-3p/NLRP3 signaling axis.

In conclusion, this study indicated ICA inhibits GC cell growth by regulating the hsa_circ_0003159/miR-223-3p/NLRP3 signaling axis. This study not only reveals the mechanism of gastric carcinogenesis but also provides potential molecular targets and therapeutic tools for its treatment.