3.1 Bulk RNA sequencing analyses reveals the role of glycosylation in the malignancy progression of gastric cancer

To investigate the specific factors influencing the progression of malignancy in gastric cancer, we utilized the dataset GSE51575 from the Gene Expression Omnibus (GEO) database (accessible at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE51575). This dataset comprises gastric cancer tissues and adjacent normal tissues from 26 patients diagnosed with gastric cancer. Following initial data preprocessing, we employed volcano plots and heatmaps to visualize the differentially expressed genes between the two groups (Fig. 1A–B). In gastric cancer tissues, the expression levels of genes such as osteopontin (SPP1), hyaluronan-degrading protein (CEMIP), cysteine protease inhibitor (CST1), signal transduction molecule (LY6E), and extracellular adhesion factor (CEACAM6) are elevated. These genes are implicated in tumor proliferation, migration, immune regulation, and other biological functions [29,30,31,32,33,34,35,36,37].

Fig. 1

Bulk RNA sequencing analyses Reveals the Role of Glycosylation in the Malignancy Progression of Gastric Cancer. A, Volcano plot showing differentially expressed genes in gastric cancer tissues and paracancerous tissues. B, Barplot of enriched KEGG pathways in gastric cancer patients. C, Heatmap demonstrating the expression levels of the top 50 differentially expressed genes. G1 represents gastric cancer tissues (n = 25), and G2 represents paracancerous tissues (n = 27).p < 0.05, |LogFC| >1 was considered significant. D-F, Gene Ontology (GO)-based functional enrichment analysis of differentially expressed genes in gastric cancer patients, including biological process (BP), cellular component (CC), and molecular function (MF) terms. G1-4, GSEA results showed that gastric cancer cells have high glycosylation modification activity

The Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analyses demonstrated that, relative to the control group, the gastric cancer patient cohort exhibited significant enrichment in functional pathways, including protein O-linked glycosylation and O-glycan processing. Additionally, we noted that the biological function of glycosylation modification was repeatedly enriched in the transcriptomic data (Fig. 1C–F). Results from Gene Set Enrichment Analysis (GSEA) further indicated elevated levels of glycosylation activity, glycosyltransferase activity, and sialic acid synthesis in gastric cancer patients (Fig. 1G1–4).

Based on a comprehensive analysis of these findings, we propose that there is an upregulation of glycosylation levels in gastric cancer tissues, contributing to tumor malignancy progression.

3.2 Sialic acid metabolism as a key factor influencing gastric cancer

To elucidate the metabolic profile of gastric cancer patients and identify metabolic pathways implicated in the progression of gastric cancer malignancy, we performed a non-targeted metabolomics analysis on gastric cancer tissues and corresponding adjacent non-cancerous tissues from 10 matched gastric cancer patients. A heatmap was generated to display the top 50 differential metabolites (Fig. 2A). Based on the results of the volcano plot of the differential metabolites (Fig. 2A)，we found that gastric cancer tissues exhibited elevated levels of acetylated metabolites, such as Acetylhomoserine, S-Acetyldihydrolipoamide, and N-Acetylglutamine, in comparison to adjacent non-cancerous tissues (Fig. 2B-D).

Fig. 2

Sialic acid metabolism as a Key Factor Influencing Gastric Cancer. In the schematic, gastric cancer tissue samples were homogenized to obtain a cell suspension, followed by lysis, centrifugation, and collection of the supernatant for detection via LC-MS/MS. A, Heatmap showing the expression levels of the top 50 differentially expressed metabolites in gastric cancer tissues (n = 10) and paracancerous tissues (n = 10).p < 0.05, |LogFC| >1 was considered as significant difference. B-C, Bar Graph of Acetylation-Related Metabolite Expression Levels E-G, Enrichment analysis of the relevant metabolic pathways based on the KEGG, SMPDB, and RaMP-DB databases.*, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. H-J, Bar chart showing expression levels of sialic acid-related metabolites

Sialic acid most common modification is the O-acetyl­ation of carbons 4, 7, 8 and 9 [38]. The difference in acetylation modifications also confers a certain level of biological functional diversity. Combining this with the bioinformatics analysis mentioned earlier, we hypothesize that high levels of acetylated metabolites, may indicate an upregulation of sialic acid metabolism within gastric cancer cells.

To undertake a comprehensive analysis of the characteristic metabolic pathways within the gastric cancer microenvironment, we conducted enrichment analyses on the differential metabolites using the KEGG, Small Molecule Pathway Database (SMPDB), and Relational Database of Metabolomics Pathways (RaMP-DB) databases. The KEGG enrichment analysis revealed that the pathways of amino sugar and nucleotide sugar metabolism were significantly upregulated. (Fig. 2E). And the metabolism of sialic acid is an important part of the overall metabolic pathway of aminoglycosides and ribonucleosides. The enrichment results of SMPDB and RaMP-DB further demonstrate the Amino Suger Metabolism, which also supports the findings of KEGG. At the same time, we also noticed the diversity of amino acid metabolism in the metabolic microenvironment, which still requires further in-depth exploration by other studies. (Fig. 2F-G).

In gastric cancer cells, glycosylation modification serves as a crucial intermediary, connecting metabolic traits to functional attributes. Notably, common glycosylation modifications encompass increased branching of N-glycans and truncation of O-glycans [39]. This observation prompted us to concentrate on a specific aspect of glycosylation: sialic acid metabolism. Traditionally, sialic acids are perceived as a composite of 2-keto acids, deoxy sugars, and amino sugars [40]. Their distinctive biological properties facilitate roles such as intracellular protein transport, mucin function, and cation transport. Various sialyltransferases are capable of mediating the sialylation of both N- and O-glycans [41]. In the context of gastric cancer, sialic acid metabolism may be attributed to the synergistic effects of multiple metabolic pathways.

Consequently, we note that the metabolomic KEGG enrichment analysis is centered on amino sugar and nucleotide sugar metabolism, whereas the bulk RNA-sequencing enrichment results highlight O-glycosylation modifications. The amino sugar and nucleotide sugar metabolism encompasses both sialic acid metabolism and sialylation, which involve both O-glycosylation and N-glycosylation. Thus, the two sets of analytical findings are not contradictory; rather, they provide complementary and deeper insights into each other.

In our analysis of the metabolic pathways involving aminoglycosides and nucleosides, we identified an upregulation of N-Acetylneuraminic acid, concomitant with a downregulation of N-Acetyl-D-mannosamine 6-phosphate and Cytidine 5’-monophosphate-N-acetylneuraminic acid (Fig. 2H–J). N-Acetylneuraminic acid, a pivotal metabolite in the sialylation processes of glycosylation pathways, plays a significant role in cell signaling, immune regulation, and the progression of malignancy in cancer cells [14, 39, 42]. Furthermore, it has been documented that elevated levels of N-Acetyl-D-mannosamine 6-phosphate can inhibit the synthesis of N-Acetylneuraminic acid [43]. This also provides indirect evidence that the metabolism of sialic acid is upregulated in gastric cancer tissues.

Our study revealed that glutamine metabolism underwent significant changes in the metabolic microenvironment of gastric cancer. (Fig.S 1A-C). As a subset of carbohydrate metabolism, glutathione metabolism plays a crucial role in mediating various redox pathways within gastric cancer cells. The potential metabolic relationship and interaction between these two pathways warrant further investigation.

3.3 Single-cell RNA sequencing analysis of communication mechanisms between subpopulations in gastric cancer

We conducted a further analysis of the data utilizing the single-cell RNA sequencing dataset GSE150290 (accessible at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE150290). During the preprocessing phase, red blood cells, as well as mitochondrial and ribosomal genes, were excluded from the dataset. Dimensionality reduction was achieved through the application of Uniform Manifold Approximation and Projection (UMAP), leading to the identification of 18 distinct cell clusters (Fig. 3A). Through the examination of differential gene expression among these clusters (Fig. 3B), we categorized them into five principal cell subpopulations: Endothelial cells (characterized by VWF, ENG, CLDN5, CDH5), Epithelial cells (characterized by KRT18, EPCAM, KRT19, PROM1, ALDH1A1), Fibroblasts (characterized by DCN, FBLN1, TAGLN, COL3A1, COL1A1, COL6A1), Lymphocytes (characterized by CD3D, CD27, CD79A, MS4A1, CD52, CD37, BLNK, MZB1, SDC1), and Monocytes and Macrophages (CD68, CD14) (Fig. 3C).

Fig. 3

Single-cell RNA Sequencing Analysis of Communication Mechanisms Between Subpopulations in Gastric Cancer. A,Uniform Manifold Approximation and Projection (UMAP) showing 18 cell clusters. B, Bubble plots demonstrating the differential expression of Marker genes between different cell clusters. C, UMAP visualization of the 18 annotated cell clusters, which were classified into five types of cellular subpopulations based on the differences in the expression of Marker genes and annotated. D, Visual analysis of the frequency and intensity of intercellular communication between different cell subpopulations based on Cell Chat. E, Bubble plot showing the landscape of different intercellular interactions.Color coding indicates communication probability, with deep blue representing low probability and deep red representing high probability. F, Bubble plot showing the intensity of MIF ligand-receptor signaling pathway.The same color gradient applies (blue = low; red = high communication probability)

In the existing literature, it is widely believed that gastric cancer cells are differentiated from epithelial cells [44, 45]. Subsequently, we conducted a CellChat analysis to evaluate the frequency and intensity of intercellular communication among various cell subpopulations (Fig. 3D). In the resulting diagram, the thickness of the arrows between cell populations indicates the strength of communication. We observed strong communication between epithelial cell subpopulations and lymphocyte subpopulations. However, the specific mode of communication remains to be further investigated. Through examining the signaling interactions mediated by ligands and receptors across distinct cell types (Fig. 3E), we identified a robust reciprocal interaction between epithelial cell subpopulations and lymphocyte subpopulations, within the MIF signaling pathway (Fig. 3F). Here, although epithelial cells were observed to interact with lymphocytes via MIF signaling, the lymphocytes did not appear to reciprocate this interaction. This phenomenon warrants further discussion. The MIF signaling pathway is implicated in several critical functions, including shielding tumor cells from macrophage-mediated cytotoxicity, facilitating tumor cell metastasis, and promoting macrophage differentiation [46,47,48].

The findings suggest that within the immune microenvironment of gastric cancer, cancer cells impose immunosuppressive effects on adjacent immune cells to sustain their proliferative and metastatic capabilities. However, the specific underlying mechanisms of intervention and effect remain to be further explored.

3.4 Upregulation of the sialic acid metabolism pathway contributes to distinct biological functional differences in gastric cancer

As also mentioned above, we regarded epithelial cells as gastric cancer cells for further analysis. We focused on the epithelial cell subgroup for further analysis to investigate the metabolic and biological functional characteristics. Using the UMAP algorithm for nonlinear dimensionality reduction with a resolution parameter set at 0.5, we identified 10 epithelial cell subsets (Fig. 4A). Differential gene expression among these clusters was analyzed using heatmaps and volcano plots (Fig. 4B–C). Our findings revealed an upregulation in the expression levels of several genes associated with tumor malignancy progression, including REG3A, FAM3D, MUC6, and S100P [49,50,51,52]. Based on the differential gene expression profiles, we categorized the ten cell clusters into high malignancy (clusters 1, 2, 3, 4, 5, 6) and low malignancy groups (clusters 0, 7, 8, 9) (Fig. 4D).

Fig. 4

Upregulation of the Sialic Acid Metabolism Pathway Contributes to Distinct Biological Functional Differences in Gastric Cancer. A, UMAP plot of re-clustered epithelial cells. These 10 cell clusters were defined by unsupervised dimensionality reduction and clustering based on single-cell gene expression profiles. B-C, Heatmap and volcano plot showing the differential genes expression between different cell clusters.p < 0.05, |LogFC| >1 is considered as significant. D, Based on differential gene expression between different cell clusters, 10 clusters were divided into high malignancy (clusters 1, 2, 3, 4, 5, 6) and low malignancy groups (clusters 0, 7, 8, 9). E, Metabolic differential analysis of high and low malignancy subpopulations using the scmetabolism package. F, Boxplots showing the expression differences of aminoglycoside and riboside metabolic pathways in high-risk and low-risk groups.G1-4, KEGG and GO enrichment analysis of functional pathways in high and low malignancy subpopulations.H1-8, Gene expression analysis of sialic acid modification-related genes detected using single-cell RNA sequencing (scRNA-seq) transcriptomic profiling.The differential expression of this gene between the HR and LR groups was visualized using a dot plot

Utilizing the scmetabolism package, we conducted a comparative analysis of metabolic differences between subgroups characterized by high and low malignancy (Fig. 4E) (red indicates high expression, and blue indicates low expression). The analysis revealed that the high malignancy subgroup demonstrated elevated levels of amino sugar and nucleotide sugar metabolism relative to the low malignancy group. Additionally, there was an upregulation in the amino sugar and nucleotide sugar metabolism metabolism pathways (Fig. 4F), corroborating previous findings in metabolomic studies.

We conducted an enrichment analysis of gastric cancer cell subgroups with varying malignancy levels using the KEGG database. The subgroups characterized by high malignancy exhibited elevated activity in pathways associated with N-Glycan biosynthesis and diverse types of N-glycan biosynthesis. Additionally, these subgroups impacted the immune microenvironment via IL-17 signaling and the regulation of Th17 cell differentiation (Figs. 4G–1). Gene Ontology (GO) enrichment analysis revealed that high malignancy gastric cancer cells demonstrated increased activity in MHC protein complex regulation (Figs. 4G–2), N-terminal glycosylation modification (Figs. 4G–3), Golgi vesicle transport, and COP I and COP II vesicle transport (Figs. 4G–3, 4). These findings further underscore the critical role of sialylation modification in the progression of gastric cancer. In the single-cell transcriptomic analysis, we primarily enriched for N-glycosylation modifications, whereas the previous findings predominantly reflected O-glycosylation (Fig. 1C–F). Given that sialylation encompasses both N-glycosylation and O-glycosylation modifications, these two distinct enrichment results are not contradictory but rather complementary, collectively revealing the functional alterations of sialic acid metabolism and sialylation in gastric cancer cells.

Enrichment analysis confirmed that glycosylation modifications play a key role in promoting malignant progression of gastric cancer. Furthermore, comparative analysis revealed that high-grade malignant gastric cancer tissues exhibited heightened biological activity across multiple functional domains compared to their low-grade counterparts, particularly manifesting enhanced energy metabolic reprogramming, dysregulated immune regulation, and intensified intracellular protein transport processes.

Furthermore, we conducted an analysis of the genes responsible for the synthesis and modification of sialic acid, as well as investigated the origins, transformation processes, and ultimate fate of sialic acid within gastric cancer tissues.

1.

Source: Our study revealed that the enzyme NANS (Figure 4H-1), a key participant in the biosynthesis of sialic acid substrates, exhibited significantly elevated expression levels in the high-risk gastric cancer cohort. This finding substantiates the upregulation of sialic acid synthesis metabolism observed in gastric cancer.

2.

Transformation: Sialic acid is converted to CMP-Neu5AC (Cytidine Monophosphate-N-acetylneuraminic acid) in the cell nucleus, and the subsequent CMP-Neu5AC participates in protein modification in the Golgi complex. Under the action of the ST3GAL1-6 family proteins, α-2,3-linked sialoglycoconjugates are generated; under the action of ST6GAL1-2 and ST6GALNAC1-6 family proteins, α-2,6-linked sialoglycoconjugates are generated; and under the action of ST8SIA1-6 family proteins, α-2,8-linked sialoglycoconjugates are generated. In our analysis, the expression of genes such as ST3GAL6 (Figure 4H-2), ST6GAL1 (Figure 4H-3), ST6GALNAC1 (Figure 4H-4), and ST6GALNAC6 (Figure 4H-5) was upregulated in the high malignancy group, while the expression of ST3GAL1 (Figure 4H-6) and ST6GALNAC4 (Figure 4H-7) was downregulated. This suggests that the high malignancy subgroup of gastric cancer cells exhibits higher sialylation modification activity. Additionally, these results also indicate a preference in pathway selection by gastric cancer cells.

3.

Fate: Compounds modified by sialic acid undergo degradation by neuraminidases in the extracellular environment, leading to the release of Neu5AC. In gastric cancer cells with high malignancy, the expression of the NEU1 (Fig. 4H-8) protein is downregulated in comparison to those with low malignancy.

In the above analysis, changes in the expression levels of a family of proteins mediating sialylation in the Golgi apparatus (ST3GAL1, ST3GAL6, ST6GAL1, ST6GALNAC1, ST6GALNAC4, ST6GALNAC6) were identified. Different genes mediate distinct modification effects, and based on the differences in their expression levels, the 10 cell clusters within gastric cancer cell subgroups were reclassified into three subgroups: G1 (cell clusters 1 and 4 with high expression of ST6GALNAC6, ST6GAL1, ST3GAL6 genes), G2 (cell clusters 2, 3, 5, and 6 with high expression of ST6GALNAC1), and G3 (cell clusters 0, 7, 8, and 9 with high expression of ST3GAL1, ST6GALNAC4) (Fig. 5A). To further explore the functional differences between these subgroups, KEGG and GO enrichment analyses revealed that G3 primarily associated with protein translation and synthesis functions, while G2 mainly involved metabolic pathways like oxidative phosphorylation, TCA cycle, and glutathione metabolism. The G1 subgroup was enriched in several immune regulation and immune cell differentiation pathways. Thus, we observed that cell subgroups with different metabolic enzyme characteristics exhibited different metabolic function preferences, which in turn promote the progression of gastric cancer malignancy (Fig. 5B).

Fig. 5

Dysregulated sialic acid metabolism underlies distinct biological function preferences in gastric cancer. A, Based on the expression differences of Golgi-mediated sialylation proteins, the gastric cancer cell subpopulations were reclassified into three subgroups (G1, G2, G3). B-C, CellChat-based visualization of intercellular interactions between different cell subpopulations, highlighting that the G1 subpopulation interacts with lymphocyte subpopulations through the MIF signaling pathway. D, KEGG and GO database-based enrichment analysis on the functional pathways of the three subgroups showing that the cellular subgroups with different metabolic enzyme profiles have different metabolic function preferences. E, The Transwell assay results (magnification 10×; scale bar = 100 μm) demonstrated the effect of sialic acid on the migration function of gastric cancer cells and were visualized using bar charts.F, The CCK-8 assay was used to analyze the effect of sialic acid on cell proliferation and viability of gastric cancer cells. G, The zebrafish fluorescence experiment was used to analyze the effect of sialic acid on the growth activity and tumorigenicity of gastric cancer cells, and the bar chart was used to visualize the average fluorescence difference of zebrafish. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Cell communication analysis of the interactions and strength of interactions between different cell subgroups showed that the G1 subgroup interacts with the lymphocyte subgroup via the MIF signaling pathway (Fig. 5C-D). Consistent with previous findings (Fig. 3F), we observed differences in MIF signaling intensity among the G1, G2, and G3 subgroups. Notably, while both G1 and G3 exhibited bidirectional MIF-mediated communication with lymphocytes, G2 showed only unidirectional signaling to lymphocytes. This variation suggests differential immunomodulatory capacities among the three subgroups. Additionally, there were also interactions among the G1, G2, and G3 subgroups. Through the coordination of multiple biological pathways, such as metabolism, immunity, and protein synthesis, these subgroups contribute to the progression of tumor malignancy.

To enhance the credibility of our findings, we conducted Transwell experiments and CCK-8 assays using gastric cancer cells (BGC-823) to investigate the effect of sialic acid (6.25 nM) on the migration and viability of gastric cancer cells. The analysis of the results revealed that the application of exogenous sialic acid led to a statistically significant increase in the number of migrating gastric cancer cells (Fig. 5E). Additionally, we observed that sialic acid promotes the proliferation of gastric cancer cells, an effect that intensifies with increasing concentrations of sialic acid (Fig. 5F). Finally, the zebrafish fluorescence detection method was used to compare the average fluorescence intensity differences between the control group and the exogenous sialic acid treatment group (Fig. 5G-1). The fluorescence intensity indicated the number of gastric cancer cells. We found that after adding sialic acid (0.05 mM), the fluorescence intensity in the zebrafish increased, and the number of gastric cancer cells rose, showing a statistical difference (Fig. 5G-2). This further confirmed the promoting effect of sialic acid on the proliferation of gastric cancer cells.

In conclusion, by analyzing key gene nodes in the sialic acid metabolism pathway, we found that genes related to sialic acid synthesis, as well as the sialyltransferase family associated with sialic acid modification, were upregulated. In contrast, protein families involved in the hydrolysis of sialylated glycoproteins were expressed in higher levels in high-malignancy gastric cancer subgroups. The combined effects of these factors led to a high level of sialylation in gastric cancer cells.

3.5 WZ35 Inhibits biological functions in gastric cancer

In our prior analysis, we identified elevated levels of sialylation modification as a distinctive metabolic characteristic of gastric cancer cells. This observation prompts the inquiry as to whether any pharmacological agent can modulate the sialylation levels in gastric cancer cells to impede their progression. Consequently, we selected the curcumin derivative WZ35 to evaluate its impact on gastric cancer cells.

CCK-8 assay results demonstrated that WZ35 inhibited the proliferation of gastric cancer cells (BGC-823) (Fig. 6A), and its inhibitory effect increased with the concentration of the drug (0 μM, 2.5 μM, 7.5 μM, 12.5 μM) (Fig. 6B). Cell apoptosis experiments showed that WZ35 also had a cytotoxic effect on gastric cancer cells (Fig. 6C).

Fig. 6

WZ35 Inhibits Biological Functions in Gastric Cancer. A, CCK-8 assay evaluating the inhibitory effect of WZ35 on gastric cancer cell proliferation. B, CCK-8 assay showing the inhibitory effect of increasing doses of WZ35 on gastric cancer cell proliferation. C, Apoptosis assay measuring the pro-apoptotic effect of WZ35 at 12.5 μM in gastric cancer cells. D, The CCK-8 experiment analyzed the changes in cell activity after the application of sialic acid, WZ35 (12.5 μM) and the combined action of sialic acid (6.25 nM) and WZ35 (12.5 μM) compared with the control group. E, The Transwell assay results (magnification 10×) demonstrated the effect of sialic acid and WZ35(12.5 μM) on the migration function of gastric cancer cells and were visualized using bar charts. F-G, The impact of sialic acid and the compound WZ35(12.5 μM) on cellular energy metabolism was assessed in gastric cancer cells using Seahorse extracellular flux analysis. H, Protein mass spectrometry analysis showing differences in expression related to the cytoskeleton, cell junctions, and other pathways between control and WZ35-treated groups(12.5 μM). I, qPCR was used to analyze the inhibitory effect of WZ35 (12.5 μM) on the mRNA expression of sialic acid-modified proteins. **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

To further investigate the relationship underlying the drug action of WZ35, we conducted the following experimental investigations. Based on the results from CCK-8 and Transwell assays, we found that WZ35 exerted inhibitory effects on the proliferation and migration of gastric cancer cells, and this inhibition could not be fully reversed even with the addition of exogenous sialic acid (Fig. 6D, E). To further investigate the effects of WZ35 and sialic acid on energy metabolism in gastric cancer cells, we conducted a Seahorse XF analysis. Four groups were set up: NC group, sialic acid group (8.25 nM), WZ35 group (12.5 μM), and combination group (WZ35 12.5 μM + sialic acid 8.25 nM). The OCR and ECAR results showed that treatment with sialic acid alone slightly upregulated mitochondrial respiration and moderately increased the glycolytic rate (ECAR). In contrast, WZ35 treatment markedly suppressed both mitochondrial respiration and glycolysis. Notably, this inhibitory effect of WZ35 could not be reversed by co-treatment with sialic acid (Fig. 6F, G). We also performed proteomic analysis comparing gastric cancer cells treated with WZ35 to control cells that were untreated (Fig. 6H). The results revealed that, after treatment with the drug, the expression of pathways related to the cell cytoskeleton and cell junctions were downregulated.

Through our experimental validation, we have demonstrated that WZ35 exhibits significant inhibitory effects on gastric cancer cell proliferation. Furthermore, we identified its suppressive activity against pan-glycosylation modifications within gastric cancer cells. Meanwhile, qPCR results demonstrated that WZ35 also exhibited inhibitory effects on key proteins involved in the previously mentioned sialylation modification pathway (Fig. 6I), the mechanism of action of WZ35 was further elucidated in this study.

3.6 WZ35 Inhibits gastric cancer progression by regulating metabolism

To elucidate the mechanisms underlying the inhibition of proliferation and metastasis of gastric cancer cells by the drug WZ35 (12.5 μM), we conducted a comprehensive metabolomic analysis on six pairs of drug-treated and control cell samples. The resulting data underwent dimensionality reduction and visualization (Fig. 7A), followed by the identification of differential metabolites between the drug-treated and control groups through t-tests and volcano plot analyses (Fig. 7B–C). Random Forest analysis was performed to identify the key metabolites [24] (Fig. 7D). Enrichment analyses utilizing the KEGG, SMPDB, and RaMP-DB databases were employed to examine alterations in the metabolic landscape of gastric cancer cells post-drug treatment. The findings revealed that WZ35 treatment led to the upregulation of pathways associated with thyroidhormonesynthesis, bile acid synthesis (Fig. 7E), branched-chain fatty acid oxidation, fatty acid metabolism (Fig. 7F). and regulation of thyroid hormone activity (Fig. 7G). Conversely, pathways related to glucose metabolism, glycolysis, aminoglycoside metabolism, nucleoside metabolism, and glutathione metabolism were downregulated.Specifically, in the metabolic enrichment analysis, the corresponding pathway enrichment results were not significantly different, ranked relatively low overall, and the results were not significantly enriched.

Fig. 7

WZ35 Inhibits Gastric Cancer Progression by Regulating Metabolism. In the schematic, WZ35-treated (12.5 μM) gastric cancer cells were collected, lysed, and centrifuged; the supernatant was then analyzed by LC-MS/MS. A, PCA for dimensional reduction and visualization of metabolic data between the NC group and WZ35-treated group, showing the relationship between sample differences. B, The differentially expressed genes in NC group and WZ35 drug-treated group were analyzed based on t-test. C, Volcano plot showing the differentially expressed genes related to sialic acid metabolism in gastric cancer tissues. p < 0.05, |LogFC|>1 was considered significant. D, Random forest model displaying key metabolites contributing to the disease(e.g., glutamine metabolites, arachidonic acid metabolites). E-G, KEGG, SMPDB, and RaMP-DB enrichment analysis of metabolic pathways, showing expchanges in the metabolic landscape after WZ35 treatment.H, Histograms showing the expression changes of sialic acid metabolites.**, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Furthermore, we analyzed the concentrations of key metabolites in the sialic acid metabolic pathway based on LC-MS/MS analysis. Our findings indicate a downregulation in the levels of sialic acid, cytidine monophosphate N-acetylneuraminic acid and Cytidine 5’-monophosphate-N-acetylneuraminic acid. These results imply that WZ35 may impede the progression of gastric cancer malignancy by inhibiting the sialic acid metabolism pathway in gastric cancer cells (Fig. 7H).