DNA-mediated ZNF649 downregulation modulates the Hedgehog signaling pathway to reduce the sensitivity of gastric cardia cancer to 5-FU

Abstract

Gastric cardia adenocarcinoma (GCA), an aggressive gastric cancer with rising incidence and poor prognosis, often requires chemotherapy and surgery. This study aimed to explore the regulatory mechanism of ZNF649 on 5-FU sensitivity in GCA. ZNF649 expression in GCA was analyzed using The Cancer Genome Atlas (TCGA) database. The relationship between ZNF649 expression and methylation level was analyzed using MethylMix and MethPrimer. GCA data was used to predict ZNF649-related signaling pathways via gene set enrichment analysis (GSEA). qPCR, western blot, and immunohistochemistry were used to detect ZNF649 expression at mRNA and protein levels. MSP was used to detect ZNF649 methylation in clinical samples. Cell viability and apoptosis were assessed by CCK8 and flow cytometry. ZNF649 was downregulated in GCA tissues and cells, negatively correlating with DNA methylation in its promoter. Overexpressing ZNF649 increased GCA cell sensitivity to 5-FU, lowering IC50 values and enhancing apoptosis. ZNF649 negatively regulated the Hedgehog pathway. Knocking down ZNF649 activated Hedgehog signaling, reducing 5-FU sensitivity. DNA methylation silences ZNF649, activating the Hedgehog pathway and weakening GCA cell sensitivity to 5-FU. Targeting ZNF649 and the Hedgehog pathway may overcome 5-FU resistance in GCA patients.

Introduction

Gastric cardia adenocarcinoma (GCA) is one of the main subtypes of gastric cancer (GC, including GCA and non-GCA), and it is also a more invasive type that mainly occurs in the transition zone between the esophagus and the stomach, with its own epidemiological, pathogenic, and clinical characteristics.^(1–3) It is worth noting that in the past decade, although the incidence and mortality rates of non-GCA have been steadily declining worldwide, the incidence rate of GCA has been increasing.^(4,5) Currently, surgery is a commonly used effective treatment method for patients with GCA.^(6,7) However, unfortunately, the postoperative recurrence and metastasis rate are often as high as 50%, resulting in poor prognosis and lower survival rate.^(5,8) Subsequently, studies have shown that chemotherapy combined with surgical treatment can reduce the postoperative recurrence and metastasis rate of patients with GCA and improve long-term survival rate, accompanied by a low incidence of adverse reactions.⁽⁹⁾ 5-fluorouracil (5-FU) is a commonly used first-line chemotherapy drug for cancer treatment, which can be used alone or in combination to treat GCA.⁽¹⁰⁾ However, the long-term application of 5-FU can lead to the rapid emergence of resistance in tumor cells. Many patients experience recurrence after receiving several courses of 5-FU chemotherapy, which has become a major problem in clinical practice. Considering the low survival rate of GCA and the resistance limitations of 5-FU in clinical applications, it is urgent to innovate new biomarkers and investigate their molecular mechanisms in GCA to improve the sensitivity of GCA to 5-FU, thus enhancing patient survival.

The zinc finger protein (ZFP) family is a large gene family that plays an essential role in the human genome. This family includes many known transcription factors, which can bind to specific nucleotide sequences upstream of genes and play precise regulatory roles in physiological processes such as transcription translation, cell differentiation, and embryonic growth.^(11,12) ZFP typically consists of a series of zinc fingers, which contain conserved cysteine and histidine ligands and bind to Zn²⁺ to form a finger-like secondary structure.⁽¹³⁾ Based on the structural characteristics of zinc finger motifs, ZFPs can be divided into seven categories: C2H2, C2C2, C2HC, C2HC4C(HD), C3H, C3HC4, and complex (containing multiple zinc finger motifs), among which, the C2H2 type is the most extensively studied type.⁽¹⁴⁾ As a member of the ZFP family, ZNF649 contains 10 C2H2-type zinc fingers and one KRAB domain, consisting of 505 amino acids. It has been noted that ZNF649 is mainly localized in the nucleus and acts as a transcriptional repressor to exert regulatory effects.⁽¹⁵⁾ For example, research has pointed out that in COS-7 cells, ZNF649 inhibits the SRE and AP-1 transcriptional activity mediated by the mitogen-activated protein kinase (MAPK) signaling pathway through KRAB domain, thereby affecting cell growth and differentiation.⁽¹⁵⁾ In addition, ZNF649 is also implicated in the malignant progression of cancer. For example, a study on breast cancer (BC) in 2022 indicated that upregulation of ZNF649 facilitates autophagy in BC cells, thereby enhancing cell sensitivity to trastuzumab.⁽¹⁶⁾ However, there have been no relevant reports on the expression of ZNF649 in GCA and its impact on the GCA cells’ sensitivity to 5-FU. Further research in this aspect can proffer more theoretical foundation for innovating ZNF469-related GCA treatment strategies.

DNA methylation is the most prominent genetic modification in the epigenome and one of the most extensively studied epigenetic regulatory mechanisms.⁽¹⁷⁾ In mammals, DNA methylation refers to the addition of a methyl group to cytosine to form 5-methylcytosine (5-mC) on cytosines within CpG dinucleotides (characterized by cytosine, guanine, and the phosphate group between them).⁽¹⁸⁾ According to relevant research statistics, the human genome has approximately 28 million CpG sites, among which CpG islands (CGIs), highly concentrated regions of CpG sites, are mainly distributed in promoter regions and regulatory regions.^(19,20) Under normal conditions, a large portion of the gene promoter region containing CGIs is in a non-methylation state. However, in many types of cancer, cancer cells can silence the expression of genes through excessive DNA methylation in the CpG island promoter region, thereby influencing the cancer process.⁽²¹⁾ For example, Han et al.⁽²²⁾ put forward in their study on GCA that the expression level of RASSF5A is frequently downregulated in GCA tissues, while its methylation level is considerably higher than that in corresponding normal tissues. Meanwhile, when the level of RASSF5A methylation is high, the survival rate of patients with GCA is low.⁽²²⁾ Similarly, the study of Guo et al.⁽²³⁾ also proved that the DNA methylation level of genes associated with GCA can affect patient survival rate. However, there is limited research on the relationship between DNA methylation and 5-FU sensitivity in GCA, with its molecular mechanism unexplored.

Based on the above views, our research focused on innovating the expression and methylation levels of ZNF649 in GCA and investigating their impact on the sensitivity of GCA to 5-FU. Meanwhile, we analyzed the relationship between ZNF649 and the Hedgehog signaling pathway in GCA, as well as their regulatory effects on GCA cells’ sensitivity to 5-FU. First, we analyzed the correlation between ZNF649 mRNA expression level and methylation level in GCA using bioinformatics analysis (TCGA database, MethylMix, and MethPrimer tools) and confirmed these results through in vitro experiments. The relationship between ZNF649 and the sensitivity of GCA to 5-FU, the relationship between ZNF649 and the Hedgehog signaling pathway, and the impact of ZNF649 and the Hedgehog signaling pathway on the sensitivity of GCA to 5-FU were also verified. In general, this project uncovered that DNA high methylation mediated low expression of ZNF649 in GCA, thereby activating the Hedgehog signaling pathway and weakening the mechanism of GCA’s sensitivity to 5-FU. This study results complement the gap of ZNF649 in the field of GCA research and are expected to provide a new direction for solving the problem of GCA drug resistance.

Materials and Methods

Patient samples

From June 2022 to July 2023, 12 pairs of human GCA patient tissue samples and adjacent tissue samples were selected and stored in liquid nitrogen for future use. This study obtained approval from the Anyang Tumor hospital Ethics Committee, and all experiments based on patient samples were approved by the committee. Furthermore, written informed consent was obtained from all patients.

Cell cultivation

In this study, human gastric mucosal epithelial cells GES-1 (BNCC337969), human embryonic kidney cells 293T (BNCC353535), and human GC cell lines AGS (BNCC338141) and HGC-27 (BNCC338546) were all purchased from BNCC (Beijing, China). AGS/5-Fu was established by culturing AGS cells for 14 days under conditions of increasing 5-Fu concentration (50–400 ‍μM). AGS/5-FU cells were cultivated in a medium containing 5-FU at a concentration of 250 ‍μM to maintain resistance. AGS cells were cultivated in an F-12K medium (Gibco, Waltham, MA) containing 10% fetal bovine serum (FBS). HGC-27 cells were cultivated in an RPMI-1640 medium (Gibco) containing 20% FBS. GES-1 cells were cultivated in a DMEM-H medium (Gibco) containing 10% FBS. 293T cells required a medium containing 2 ‍mM l-glutamine. All cells were cultured in a cell culture incubator at a temperature of 37°C, humidity of about 90%, and 5% CO₂.

Cell transfection

The ZNF649 coding sequence was synthesized and cloned into the pcDNA3.1 vector by GenePharma (Shanghai, China) to construct oe-ZNF649 plasmid, with empty pcDNA3.1 plasmid as a negative control (oe-NC). si-ZNF649 and si-NC were designed and synthesized by RiboBio (Guangzhou, China). The transfection reagent Lipofectamine 3000 (Invitrogen, Waltham, MA) was used to transfect plasmids into cells. In simple terms, cells were first cultured to 70–90% confluence, and then the reaction mixture was prepared according to the requirements and mixed with the cell culture medium. After incubation at room temperature for 5 ‍min, DNA-liposome complexes were added and the cells were incubated at 37°C for 48 ‍h. Transfected cells were then analyzed for subsequent experiments. Transfection was performed for 48 ‍h, and cells were collected for subsequent experiments.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR)

Total RNA was extracted from cells using TRIzol (Invitrogen). RNA concentration and purity were determined using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). Total RNA was then synthesized into cDNA using PrimeScript^TM RT reagent Kit (Takara Bio, Kusatsu, Japan). Next, qRT-PCR was performed using AceQ qPCR SYBR Green Master Mix (Vazyme, China) on the Applied Biosystems 7500 rapid real-time fluorescence quantitative PCR system (Thermo Fisher Scientific). GAPDH was applied as an internal reference, with the relative expression level of the target gene mRNA calculated using the 2^−ΔΔCt method. The primer sequences used were purchased from Beijing Tsingke Biotech Co., Ltd (Beijing, China). Table 1 is a list of primer sequences.

Table 1.

Primer sequence

Name	Primer sequence (5'→3')
ZNF649 (NM_023074.4)	F: AGGATTTAGAAGAAATTAAAGATGA R: TAATAAAAACATCAAATTTACCAAC
GAPDH (NM_001256799.3)	F: GATTCCACCCATGGCAAATTC R: CTGGAAGATGGTGATGGGATT

Western blot (WB)

Cells were lysed using RIPA lysis buffer (Beyotime, Shanghai, China) containing protease inhibitors and phosphatase inhibitors. The lysates were placed on ice for 10 ‍min and then centrifuged at 4°C and 12,000 ‍rpm for 10 ‍min to collect the supernatant. The protein concentration was determined using Pierce^TM BCA Protein Assay Kits (Thermo Fisher Scientific). The protein sample containing 1× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) protein loading buffer was denatured at 100°C for 7 ‍min and then subjected to electrophoresis using 10% SDS-PAGE. After electrophoresis, the protein was transferred to the polyvinylidene fluoride membrane, which was incubated in 5% skim milk for 1 ‍h, Tris-Buffered Saline with Tween 20 (TBST) was washed 3 times, and then incubated with primary antibody at 4°C overnight. After rinsing with TBST, the protein was incubated with the secondary antibody for 1 ‍h. After washing with TBST again, the protein was observed using BeyoECL Plus (Ultra-sensitive ECL kit) (P0018M; Beyotime). Finally, the band images were visualized and captured using the ChemiScope 6000 chemiluminescence imaging system (Clinx, Shanghai, China). Primary antibodies included ZFN649 antibody (PA5-31905; Thermo Fisher Scientific), SMO antibody (PA5-76145; Thermo Fisher Scientific), GLI1 antibody (ab134906; Abcam, Cambridge, UK), Shh antibody (ab308225; Abcam), GAPDH antibody (ab9485; Abcam). The second antibody was goat anti-rabbit IgG H&L (HRP) (ab205718; Abcam).

Methylation-specific polymerase chain reaction (MSP)

Genomic DNA was extracted from GCA tissues, normal tissues, GCA cell lines, and normal cell lines using the proteinase K digestion method. Next, to detect the methylation level of the ZNF649 gene, genomic DNA was treated with the Epitect Fast Bisulfite Conversion Kit (Qiagen, Hilden, Germany). After treatment, the unmethylated cytosine was converted to uracil by treatment with sodium bisulfite, while the methylated cytosine remained unchanged. Based on the potential differences in DNA sequences between methylation and non-methylation after treatment with bisulfite, we designed two pairs of primers for PCR and used gel electrophoresis to analyze the level of ZNF649 methylation. The primers for methylation were 5'-AGGATTTAGAAGAAATTAAAGACGA-3' and 5'-TAATAAAAACATCAAATTTACCGAC-3'. The primers for non-methylation were 5'-GAGGATTTAGAAGAAATTAAAGATGA-3' and 5'-TTAATAAAAACATCAAATTTACCAAC-3'. The sequences were purchased from Beijing Tsingke Biotech Co., Ltd. In addition, genomic DNA treated with CpG methyltransferase (Sss I) (New England BioLabs, Ipswich, MA) was utilized as a positive control, with sterile water as a negative control.

CCK-8 assay

A total of 5 × 10³ AGS cells were seeded into a 96-well plate. After cell attachment, culture medium containing different concentrations of 5-FU (0, 50, 100, 200, and 400 ‍μM) was added to the wells and incubated for 48 ‍h at 37°C in a 5% CO₂ incubator. After incubation with 10 ‍μl CCK-8 solution (Dojindo Molecular Technologies, Kumamoto, Japan) for 2 ‍h, the optical density (OD) value was detected at 450 ‍nm using a microplate reader. The half-inhibitory concentration was calculated using GraphPad Prism 9 (GraphPad Software, Boston, MA).

Flow cytometry

The apoptosis rate was analyzed. Experiments were performed using Annexin V-FITC/PI Apoptosis Kit (MultiSciences Biotech, Zhejiang, China). In short, the collected cells were centrifuged at 1,000 ‍rpm for 5 ‍min in a centrifuge, with the supernatant discarded. The cells were washed once with phosphate-buffered saline (PBS) and gently resuspended for counting. Then, 1 × 10⁶ cells were resuspended in 500 ‍μl 1× Binding Buffer solution, and 5 ‍μl AnnexinV-FITC and 10 ‍μl PI were added sequentially. After gentle vortex mixing, the cells were incubated at room temperature in the dark for 5 ‍min. Cell analysis was performed using the NovoCyte flow cytometer system (Agilent, Santa Clara, CA), and the apoptosis rate was calculated and analyzed using Flowjo_V10 software.

Data processing and analysis

Statistical analysis of the data was carried out on GraphPad Prism software (GraphPad Software). The data was presented as mean ± SD. The t test and one-way analysis of variance (ANOVA) were applied to determine the significance of differences in two groups and multiple groups, respectively. Statistical analysis was performed using data collected from at least three independent experiments. A p value<0.05 was considered statistically significant.

Bioinformatics analysis

The mRNA expression data of GCA were downloaded from the TCGA database (normal: 12, tumor: 90). The read mRNA data were normalized and analyzed for differences (|logFC|>0.585, FDR<0.05) using the edgeR package. Subsequently, the normalized expression profiles of the target genes were extracted and plotted as violin maps. Then, the expression level of ZNF649 was divided into high- and low-expression groups for gene set enrichment analysis (GSEA) to explore ZNF649-related signaling pathways. In addition, based on the TCGA database, the methylation level data of ZNF649 in GCA samples and normal samples were downloaded (normal: 12, tumor: 100), with data standardized by using the limma package. The correlation between ZNF649 mRNA expression level and methylation level, as well as the gene sequence of the ZNF649 promoter region (2,000 ‍bp), were analyzed and predicted using the MethylMix package (Cor<−0.3) and the MethPrimer website.

Results

ZNF649 exhibits high methylation level and low expression level in GCA

To clarify the expression of ZNF649 in GCA, we conducted relevant research. First, the mRNA expression data of ZNF649 in GCA samples and normal samples were downloaded from the TCGA database and subjected to differential expression analysis after normalization. The results uncovered that compared with normal samples, the mRNA level of ZNF649 in GCA samples was considerably decreased (Fig. 1A). Subsequently, we downloaded the methylation level data of ZNF649 in GCA samples and normal samples from the TCGA database, and analyzed and predicted the correlation between ZNF649 mRNA expression level and methylation level, as well as the gene sequence of the ZNF649 promoter region (2,000 ‍bp), to determine whether ZNF649 was a driver gene in GCA. The results unearthed a significant negative correlation between the expression level of ZNF649 and its methylation level. The ZNF649 promoter region contained two CGIs (Fig. 1B and C). These results indicated that in GCA, ZNF649 underwent high methylation, resulting in a decrease in its mRNA expression level.

Fig. 1.

Bioinformatics prediction and experimental validation of the expression level and methylation level of ZNF649 in GCA. (A) The TCGA predicted the mRNA expression level of ZNF649 in GCA. (B) The MethylMix package analyzed the correlation between ZNF649 expression and methylation level in GCA. (C) The MethPrimer website predicted the promoter region sequence of ZNF649. (D) qRT-PCR detected the expression of ZNF649 in cancer-adjacent normal samples and GCA samples. (E) MSP experiment detected the methylation level of ZNF649 in cancer-adjacent normal samples and GCA samples. M represents the primer for methylation; U represents the primer for non-methylation; NC represents negative control; PC represents positive control; T represents GCA samples; N represents cancer adjacent samples. (F) qRT-PCR measured the mRNA expression levels of ZNF649 in normal cell lines and GCA cell lines. (G) WB measured the protein expression level of ZNF649 in normal cell lines and GCA cell lines. (H) MSP detected the methylation level of ZNF649 in normal cell lines and GCA cell lines. (I) qRT-PCR analyzed the mRNA expression levels of ZNF649 in normal cell lines and GCA cell lines after 5az treatment. (J) WB detected the protein expression level of ZNF649 in normal cell lines and GCA cell lines after 5az treatment. Data are presented as mean + SD, and the differences are statistically significant (*p<0.05).

To further validate the accuracy of bioinformatics analysis, we conducted in vitro experiments. First, the expression of ZNF649 in the collected 12 pairs of clinical samples (GCA samples and cancer-adjacent samples) was measured by utilizing qRT-PCR. The results uncovered that compared to cancer-adjacent samples, the expression level of ZNF649 was considerably decreased in GCA samples (Fig. 1D). Then, we detected the methylation level of ZNF649 in clinical samples using MSP experiments, finding that the level of ZNF649 in GCA samples was considerably elevated than that in cancer-adjacent samples (Fig. 1E). In addition to gastric cardia tissue, we also conducted relevant experimental tests on the expression of ZNF649 at the cellular level. First, we utilized qRT-PCR and WB to detect and analyze the expression of ZNF649 in normal cell lines and GCA cell lines, unearthing that both the mRNA expression level and protein expression level of ZNF649 were considerably higher in normal cell lines than in GCA cell lines (Fig. 1F and G). Next, we detected the levels of ZNF649 in normal cell lines and GCA cell lines using MSP experiments, which demonstrated that compared to normal cell lines, the methylation level of ZNF649 in GCA cell lines was considerably increased (Fig. 1H). In addition, we analyzed the expression of ZNF649 in the above cell lines after treatment with 5az (a methyltransferase inhibitor) by qRT-PCR and WB. The results uncovered that there were no significant changes in the mRNA and protein levels of ZNF649 in normal cell lines after 5az treatment, while both the mRNA and protein levels of ZNF649 in GCA cell lines were considerably increased after 5az treatment (Fig. 1I and J). It should be noted that considering the more significant level changes in the GCA cell line AGS, subsequent experiments in this study were conducted using the AGS cell line. Overall, all the above results indicated that in GCA, ZNF649 underwent high methylation, resulting in lower mRNA expression levels and protein expression levels.

ZNF649 facilitates the sensitivity of GCA to 5-FU

Considering the differential expression of ZNF649 in GCA, we investigated the impact of ZNF649 on the sensitivity of GCA to 5-FU, followed by relevant experiments. First, the ZNF649 overexpression plasmid was transfected into AGS cells to construct the normal expression group and ZNF649-overexpressed group. Transfection efficiency was verified by qRT-PCR and WB experiments, which revealed that compared to the normal expression group, the mRNA and protein levels of ZNF649 were considerably increased in the ZNF649-overexpressed group, indicating successful transfection of plasmid (Fig. 2A and B). Subsequently, two groups of cells were treated with different concentrations of 5-FU solution (0, 50, 100, 200, and 400 ‍μM), with the IC₅₀ values of the two groups of cells detected and calculated using the CCK8 assay. The results demonstrated that the IC₅₀ value of the control group cells was 217.8, while the IC₅₀ value of cells in the ZNF649-overexpressed group was 87.60, which was considerably lower than that of the ZNF649-normally-expressed group (Fig. 2C), suggesting that the sensitivity of AGS cell line to 5-FU was enhanced when ZNF649 was overexpressed. In addition, we also utilized flow cytometry to detect the apoptosis of the two groups. finding that the apoptosis rate in the ZNF649-overexpressed group was considerably higher than that in the ZNF649-normally-expressed group (Fig. 2D). In general, these experimental results indicated that high expression of ZNF649 in GCA cells can increase cell sensitivity to 5-FU and enhance apoptosis ability.

Fig. 2.

Experimental verification of the impact of ZNF649 on 5-FU resistance in GCA (A) qRT-PCR detected mRNA expression levels in AGS cells with normal expression or overexpression of ZNF649. (B) WB analyzed protein expression levels of ZNF649 in AGS cell lines with normal or overexpression of ZNF649. (C) CCK8 detected and calculated the IC₅₀ values of the two cell groups. (D) Flow cytometry measured the apoptosis rate of two cell groups. Data are presented as mean + SD, and the differences are statistically significant (*p<0.05).

ZNF649 modulates the Hedgehog signaling pathway to affect the sensitivity of GCA to 5-FU

This project aimed to further interpret the regulatory mechanism of ZNF649 on the sensitivity of GCA to 5-FU. Therefore, based on the expression data of ZNF649 in the TCGA-GCA dataset, GSEA revealed that ZNF649 was enriched in the Hedgehog signaling pathway, indicating a correlation between ZNF649 and the Hedgehog signaling pathway (Fig. 3A). Therefore, we speculated that ZNF649 affects the sensitivity of cells to 5-FU in GCA by modulating the Hedgehog signaling pathway. To validate this hypothesis, AGS cells were cultivated under conditions of increasing 5-Fu concentrations (50–400 ‍μM) for 14 days to establish AGS/5-Fu cells. AGS/5-Fu cells were then cultured in a medium containing 250 ‍μM 5-Fu to maintain resistance. Based on AGS and AGS/5-Fu cells, experimental groups were constructed, including the normal expression group (sh-NC + DMSO), ZNF649-silenced group (sh-NF649 + DMSO), and ZNF649 silenced and GANT61 co-treated group (sh-ZNF649 + GANT61, GANT61 is a Hedgehog signaling pathway inhibitor). Subsequently, the protein expression levels of ZNF649 and Hedgehog signaling pathway-related proteins (SMO, GLI1, and Shh) in cells of each treatment group were detected and analyzed using WB experiments, which unearthed that the expression of ZNF649 was considerably downregulated in the ZNF649-silenced group, while the expression levels of Hedgehog signaling pathway-related proteins were upregulated. In the ZNF649-silenced combined with GANT61 treatment group, the expression levels of Hedgehog signaling pathway-related proteins were partially restored (Fig. 3B), indicating that ZNF649 negatively regulated the Hedgehog signaling pathway. Next, we employed CCK8 assay and flow cytometry to detect the viability and apoptosis ability of cells in each group, finding that compared with the normal expression group, the cell viability of AGS and AGS/5-FU cells in the ZNF649-silenced group was considerably elevated, while the apoptosis rate was considerably decreased. In the co-treated group, the cell viability and apoptosis rate of AGS and AGS/5-FU cells were restored to varying degrees (Fig. 3C and D). Overall, these experimental results indicated that in GCA cells, ZNF649 downregulated the Hedgehog signaling pathway, thereby increasing cell sensitivity to 5-FU, reducing cell viability, and increasing the apoptosis rate.

Fig. 3.

Experimental verification of the relationship between ZNF649 and the Hedgehog signaling pathway, as well as their impact on 5-FU resistance in GCA. (A) GSEA analyzed ZNF649-related signaling pathways. (B) WB detected the protein expression levels of ZNF649 and Hedgehog signaling pathway-related proteins SMO, GLI1, and Shh in cells from each treatment group. (C) CCK-8 assay measured cell viability in the three groups. (D) Flow cytometry detected the apoptosis rate of cells in each treatment group. Data are presented as mean + SD, and the differences are statistically significant (*p<0.05).

Discussion

In our project, ZNF649 was first observed to be downregulated in GCA tissues and cell lines, exhibiting a negative correlation with its DNA methylation levels. Cell experiments demonstrated that the expression of ZNF649 enhanced the sensitivity of GCA to 5-FU. In addition, using the bioinformatics tool, we predicted the correlation between ZNF649 and the Hedgehog signaling pathway and further verified their regulatory relationship, as well as the impact of the ZNF649/Hedgehog signaling pathway regulatory axis on the sensitivity of GCA to 5-FU. Therefore, this project revealed that ZNF649 was downregulated by DNA methylation, thereby activating the Hedgehog signaling pathway and reducing the sensitivity of GCA cells to 5-FU.

The occurrence and proliferation of cancer are modulated by epigenetic modifications, which have been identified as a focus of GC research.⁽²⁴⁾ DNA methylation, as an important epigenetic modification, has strong stability and can be inherited by progeny cells during cell division.⁽²⁵⁾ In recent years, DNA methylation has gradually become a research hotspot in the field of cancer as an epigenetic modification, and abnormal DNA methylation can be identified in various cancers.^(24,26) The promoter regions of various tumor suppressor genes (TSGs) are high-risk regions for DNA methylation in cancer. Due to the high methylation of TSG, TSG cannot play its anticancer role, which is one of the molecular mechanisms of DNA methylation in promoting cancer progression.⁽¹⁷⁾ In this investigation, we unearthed that ZNF649 was downregulated in GCA cells. ZNF649 is a type of ZFP, which usually exhibits anti-cancer effects in tumors. Therefore, we speculated that ZNF649 may be implicated in the upregulation of methylation in GCA, limiting the expression of this gene. Given this, we utilized bioinformatics tools to predict the expression level of ZNF649, finding that ZNF649 was negatively correlated with the level of DNA methylation, and there were two CGIs of DNA methylation in the promoter region of ZNF649. Subsequently, we detected significant downregulation of ZNF649 mRNA and protein expression levels, as well as a significant increase in methylation levels, in GCA tissues and cells by utilizing qPCR, WB, immunohistochemistry, and MSP experiments. Collectively, in GCA, ZNF649 exhibited a high methylation, resulting in a lower expression level, which is consistent with the mainstream model of DNA methylation in the process of tumor development. In fact, similar phenomena have been observed in different types of cancers, such as GC, endometrial cancer, thyroid cancer, esophageal squamous cell carcinoma, etc., where high methylation levels of TSGs lead to downregulation of expression, thus affecting the growth and proliferation of cancer cells.^(27–31)

In this project, the role of ZNF649 in GCA and its related regulatory mechanisms were fully discussed. In a study on colorectal cancer (CRC), Chauvin et al.⁽³²⁾ pointed out that when CRC cells were treated with 5-FU resistance, quantitative proteomics and transcriptomics analysis revealed a significant downregulation of the ZNF649 gene in cancer cells. The expression of ZNF649 is negatively correlated with the resistance of CRC cells.⁽³²⁾ Our results are in line with that conclusion. ZNF649-overexpressed cells and ZNF649-normally-expressed cells were constructed based on the AGS cell line. The two groups of cells were treated with different concentrations of 5-FU. The IC₅₀ value of ZNF649-overexpressed cells was remarkably lower than that of ZNF649-normal-expressed cells, and its apoptosis ability was substantially enhanced. Consequently, overexpression of ZNF649 facilitated the sensitivity of GCA cells to 5-FU. It is worth noting that in the study of Chauvin et al.,⁽³²⁾ the relationship between ZNF649 and 5-FU sensitivity only remains at the level of omics analysis. However, this project confirmed that ZNF649 positively modulated the 5-FU sensitivity of cells at the cellular level. In addition, this project further explored the regulatory mechanism of ZNF649 on cell sensitivity to 5-FU in GCA. Based on the analysis of ZNF649 in the TCGA-GCA database, we observed a remarkable correlation between ZNF649 and the Hedgehog signaling pathway. Cell experiments indicated that silencing ZNF649 could activate the Hedgehog signaling pathway, thereby reducing the sensitivity of GCA cells to 5-FU. The role of the Hedgehog signaling pathway proposed in this project in drug sensitivity was consistent with previous results in hepatocellular carcinoma (HCC) research. Wang et al.⁽³³⁾ uncovered in their study that the Hedgehog signaling pathway in HCC can facilitate sensitivity to sorafenib in HCC. After using the Hedgehog signaling inhibitor GANT61, cell viability can be effectively inhibited, thus improving resistance.⁽³³⁾ The Hedgehog signaling pathway may have similar roles in different types of cancer.

In conclusion, this investigation revealed a new mechanism: DNA methylation mediated the downregulation of ZNF649 to activate the Hedgehog signaling pathway, thus reducing the sensitivity of GCA to 5-FU. This finding suggested that targeting ZNF649 or the Hedgehog signaling pathway may facilitate the sensitivity of GCA to 5-FU, improve chemotherapy efficacy, alleviate adverse prognosis, and increase the patient survival rate. Unfortunately, this project did not conduct relevant animal experiments, which is a research direction that needs to be supplemented and improved in the future. Animal experiments can better simulate the physiological environment of GCA, evaluate the role of ZNF649 and Hedgehog signaling pathways in vivo, and validate the effectiveness and safety of targeting these molecules for therapeutic strategies. Further research will aid in exploring the mechanism of ZNF649 and the Hedgehog signaling pathway in GCA, proffering a more reliable scientific basis for the development of new treatment strategies.

Author Contributions

GX conceived of the study, and participated in its design and interpretation and helped to draft the manuscript. BL and XL participated in the design and interpretation of the data and drafting/revising the manuscript. YW and ZL performed the statistical analysis and revised the manuscript critically. All the authors read and approved the final manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not for profit sectors.

Ethics Approval and Consent to Participate

This study was approved by the Anyang Tumor Hospital Ethics Committee, approval date: 28-08-2024, approval number: 2024WZ17K01.

Data Availability Statement

The data and materials in the current study are available from the corresponding author on reasonable request.

Consent to Participate Statement

Patient consent was not required in accordance with local or national guidelines.

Conflict of Interest

No potential conflicts of interest were disclosed.