Abstract Background Enhanced glycolytic levels in cancer cells are a common characteristic of many cancer types. Modulation of glycolytic metabolism is crucial for enhancing the efficacy of cancer therapy. The specific role of 6-methoxyflavone in regulating glycolytic metabolism in cancer cells remains unclear. This study aimed to elucidate the impact of 6-methoxyflavone on glycolytic metabolism in cervical cancer cells and its clinical relevance. Methods The tandem mass tag (TMT) proteomic analysis was used to identify significantly enriched biological processes and pathways in HeLa cells after treatment with 6-methoxyflavone. Additionally, the differential expression of glycolysis-related proteins was validated using parallel reaction monitoring (PRM) proteomics. Untargeted and targeted metabolomics analyses were used to identify differentially expressed glycolysis-related metabolites. Furthermore, alternative splicing, new transcripts, and domain analyses were used to detect the effects of 6-methoxyflavone on the structures of glycolysis-related genes and proteins. Subcellular localization, molecular docking, and non-covalent interaction analyses were used to detect the subcellular localization, affinity of 6-methoxyflavone for glycolysis-related proteins, and sites of non-covalent interactions. Clinical characteristics and immunological correlation analyses were used to elucidate the relationships between glycolysis-related genes and clinicopathological characteristics, survival, prognosis, and immune-related indicators of patients with cervical cancer. Finally, glycolysis stress tests and enzyme activity assays were used to verify the effect of 6-methoxyflavone on glycolysis in HeLa cells. Results TMT and PRM proteomics, as well as untargeted and targeted metabolomics results, showed that 6-methoxyflavone downregulated the expression levels of glycolysis-related proteins and metabolites in HeLa cells, and that the structures and functions of glycolysis-related genes and proteins in the cytoplasm underwent changes. 6-Methoxyflavone had a good affinity for nine glycolysis-related proteins, all of which had non-covalent interaction sites. Clinical characteristics and immune correlation analyses showed relationships between 6-methoxyflavone and five clinical characteristics, survival prognosis, and four immune-related indicators in patients with cervical cancer. After treatment with 6-methoxyflavone, the basal glycolytic level, maximum glycolytic capacity, and glycolytic reserve of HeLa cells were downregulated. Additionally, 6-methoxyflavone inhibited the activity of pyruvate kinase. Conclusion 6-Methoxyflavone inhibited energy metabolism in HeLa cells through the glycolysis pathway. 6-Methoxyflavone may be related to five clinical characteristics, prognosis, tumor microenvironment, immune cells, immune checkpoints, and immunotherapy efficacy in patients with cervical cancer. Supplementary Information The online version contains supplementary material available at 10.1186/s12885-025-14133-9. Keywords: HeLa cells, Glycolysis, Energy metabolism, 6-methoxyflavone, Cervical cancer Introduction Recently, the incidence and mortality of cervical cancer have continuously increased [[26]1]. According to the Global Cancer Observatory 2022 ([27]https://gco.iarc.who.int/today), the number of new cases of cervical cancer and deaths worldwide in 2022 was 662,301 and 348,874, respectively. Cervical cancer remains a major threat to the health of women. In the treatment of cervical cancer, traditional radiotherapy and chemotherapy face problems such as treatment resistance [[28]2] and serious side effects [[29]3]. Cellular immunotherapy has limited efficacy owing to the presence of an immunosuppressive microenvironment in cancer tissues, long production cycles, and severe adverse reactions [[30]4]. Targeted therapeutic drugs and immune checkpoint inhibitors are often used in patients positive for specific targets; however, their applicability is limited. Hence, in-depth research on drugs for cervical cancer treatment is urgently needed. Glycolysis reprogramming is an important feature of cancer [[31]5], closely related to the proliferation, migration [[32]6], invasion [[33]7], metastasis [[34]8], treatment sensitivity [[35]9], treatment resistance [[36]10], and patient prognosis [[37]8] of cancer cells. Warburg discovered that cancer cells primarily rely on glycolysis for energy [[38]11]. Tumor glycolysis is considered a promising new target for the development of anticancer drugs [[39]5]. Among the 175 approved small-molecule drugs for treating tumors, drugs from natural sources account for nearly half [[40]12]. 6-Methoxyflavone is extracted from the natural Chinese herbal medicine Imperata cylindrica [[41]13]. Previous studies of Imperata cylindrica have shown a wide range of effects, including anti-inflammatory [[42]14], antifungal [[43]15], anticancer [[44]16], antioxidant [[45]17], and anti-fibrotic [[46]18] properties. This herbal medicine has shown therapeutic efficacy against various types of cancer, including cervical [[47]16], colorectal [[48]19], oral [[49]20], colon, liver, and breast [[50]21] cancers. Furthermore, existing studies have indicated that 6-methoxyflavone possesses anti-inflammatory [[51]22], anticancer [[52]23], and immunomodulatory [[53]24] properties, and is effective in photodynamic therapy for cancer [[54]25], as well as in alleviating the side effects of chemotherapy [[55]26]. 6-Methoxyflavone has been found to have therapeutic effects on cervical cancer, gastric cancer, and melanoma [[56]23]. However, the specific role of 6-methoxyflavone in regulating glycolytic metabolism in cancer cells remains unclear. Cancer cells primarily obtain their energy through the process of glycolysis [[57]27]. Enolase 3 (ENO3) is a glycolysis-related protein that is notably upregulated in colorectal cancer (CRC) and has been implicated in the progression of CRC by increasing glycolytic activity [[58]28]. Pyruvate kinase (PKM) plays a crucial role in the glycolytic pathway and contributes to tumorigenesis, metastasis, stemness maintenance, and macrophage polarization in hepatocellular carcinoma through its role in promoting glycolysis [[59]29]. Lactate dehydrogenase A (LDHA), another glycolytic enzyme, has been shown to facilitate the proliferation, invasion, and metastasis of pancreatic cancer cells by intensifying glycolytic processes [[60]30]. Other glycolysis-related proteins, such as lactate dehydrogenase B (LDHB) [[61]31], fructose-bisphosphate aldolase A (ALDOA) [[62]32], glucose- 6-phosphate isomerase (GPI) [[63]7], phosphoglycerate kinase 1 (PGK1) [[64]33], phosphoglucomutase- 1 (PGM1) [[65]34], and triosephosphate isomerase (TPI1) [[66]7], also play significant roles in cancer progression by enhancing glycolysis. Several small molecules have been identified that exhibit potent inhibitory effects on glycolysis-related proteins [[67]7, [68]35–[69]37] and possess anticancer properties. Thus, targeting tumor energy metabolism through the strategic inhibition of glycolysis-related proteins may represent a promising approach for the development of novel anticancer therapies [[70]35]. Furthermore, glycolytic metabolites are crucial in the advancement of cancer by influencing the tumor microenvironment. Pyruvate, a key glycolytic metabolite, is derived from several upstream metabolites in the glycolytic pathway, including D-glucose 1-phosphate, glucose 6-phosphate, and D-fructose 6-phosphate. Exogenous pyruvate has been shown to downregulate the expression of histone genes and impede the proliferation of cancer cells [[71]38]. In breast cancer, these cells depend on environmental pyruvate to establish and modify the metastatic niche [[72]39]. Hyperpolarized 13 C pyruvate functions as a non-invasive biomarker for the detection and grading of prostate cancer [[73]40]. The blockade of pyruvate entry into the mitochondria enhances the sensitivity of prostate cancer to both 18 F-PET/CT detection and radiotherapy [[74]41]. Therefore, glycolytic metabolites may provide a critical target for cancer diagnosis and treatment [[75]40, [76]41]. This context led us to hypothesize that 6-methoxyflavone may exert its effects through glycolysis-related proteins or metabolites to inhibit energy metabolism in HeLa cells. The objective of this study was to elucidate the potential role and mechanism of action of 6-methoxyflavone in the glycolytic energy metabolism of HeLa cells. Materials and methods 6-Methoxyflavone and cell culture 6-Methoxyflavone powder was purchased from Weikeqi Biological Co. Ltd. (Chengdu, China). Using ultrasound, dimethyl sulfoxide (DMSO) was used to dissolve 6-methoxyflavone. HeLa and SiHa human cervical cancer cell lines, as well as the HaCaT human normal cell line, were obtained from the School of Basic Medicine, Peking Union Medical College (Beijing, China). HeLa cells were cultured in Dulbecco’s modified Eagle’s medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum. SiHa and HaCaT cells were cultured in modified Eagle’s medium (HyClone) supplemented with 10% fetal bovine serum. The temperature, relative humidity, and carbon dioxide concentration for all cell cultures were 37 ℃, 95%, and 5%, respectively. Cell proliferation and toxicity assay Single-cell suspensions were seeded in 96-well plates for 24 h. Subsequently, the HeLa cells were treated with six concentrations of 6-methoxyflavone (0.16% DMSO, 10 μM, 20 μM, 40 μM, 80 μM, and 100 μM) for 24, 48, and 72 h. Following this, the cell counting kit 8 (CCK8) (MeilunBio, Dalian, Liaoning, China) was used to assess the toxicity of 6-methoxyflavone on HeLa cells. The optical density values were detected at 450 nm using a microplate reader MB580 (Huisong Heales, Shenzhen, Guangdong, China). The half-maximal inhibitory concentration (IC50) values were calculated using GraphPad Prism software ([77]www.graphpad.com). TMT proteomics and its functional and pathway enrichment analysis Tandem Mass Tag (TMT) technology was used for quantitative proteomic analysis. This methodology utilizes isotopic reagents to label the amino groups at the termini of peptides or the lysine side chains of proteins. By integrating liquid chromatography with mass spectrometry, researchers are able to conduct simultaneous comparisons of protein expression levels across multiple samples. The six-plex TMT high-throughput proteomics method allows for three biological replicates within a single group, ensuring reliable statistical analysis for quantitative measurements [[78]42]. In this study, six-plex TMT proteomic analysis was utilized to identify differentially expressed proteins. HeLa cells were treated with 0.16% DMSO and 6-methoxyflavone (65 μM) for 48 h. The control group and the treatment group were each subjected to three replicates. Cell lysis and protein extraction were conducted using a buffer composed of 4% sodium dodecyl sulfate, 100 mM Tris hydrochloride, and 1 mM dithiothreitol, with the pH adjusted to 7.6. Protein digestion was carried out with trypsin, adhering to the filter-aided sample preparation protocol developed by Matthias Mann [[79]43]. A total of 100 μg of peptide mixture from each sample was labeled utilizing TMT Label Reagent (41 μL, 19.53 μg/μL) (Thermo Fisher Scientific, Madison, WI, USA). The reaction was allowed to incubate for one hour at room temperature. Liquid chromatography-mass spectrometry/mass spectrometry analysis was performed using a Q-Exactive mass spectrometer (Thermo Fisher Scientific) that was coupled to an Easy nLC chromatograph system (Proxeon Biosystems, now Thermo Fisher Scientific, Madison, WI, USA) at Applied Protein Technology Biotech Co., Ltd. (Shanghai, China). The detection mode was positive ion mode with a scanning range of parent ions from 300 to 1800 m/z. The maximum ion time was 50 ms, and the dynamic exclusion was set at 60.0 s. The automatic gain control target was 1e6. The screening criteria for the significantly differentially expressed proteins in this study were a fold change > 1.2 or < 0.84, and p < 0.05. This screening criterion was derived from the published articles [[80]44–[81]46]. Blast2GO [[82]47] software was used for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) [[83]48] analyses of differentially expressed proteins. Differential GO and KEGG items were identified at p < 0.05. Analyses of untargeted and targeted metabolomics Glycolysis is a crucial process in cellular energy metabolism. 6-Methoxyflavone significantly inhibits glycolysis in HeLa cells. After 6-methoxyflavone treatment, we used high-resolution untargeted and targeted metabolomics to detect the expression of glycolysis-related metabolites in HeLa cells. Untargeted metabolomics was detected using the positive and negative ion modes of electrospray ionization. Liquid chromatography-mass spectrometry/mass spectrometry analysis was performed using a Q-Exactive mass spectrometer (Thermo Fisher Scientific) that was coupled to a Vanquish UHPLC chromatograph system (Thermo Fisher Scientific) at Applied Protein Technology Biotech Co., Ltd. (Shanghai, China). The detection range for the first-level mass-to-charge ratio was 80–1200 Da, with a resolution of 60,000 and a scanning accumulation time of 100 ms. The secondary mass spectrometry utilized a segmented acquisition approach, encompassing a scanning range from 70 to 1200 Da, a resolution of 30,000, a scanning accumulation time of 50 ms, and a dynamic exclusion time of 4 s. Targeted metabolomics were detected in positive and negative switch modes. Liquid chromatography-mass spectrometry/mass spectrometry analysis was performed using a 6500 + QTRAP mass spectrometer (AB Sciex, Pte. Ltd., USA) that was coupled to a 1290 Infinity chromatograph system (Agilent Technologies, Winooski, VT, USA) at Applied Protein Technology Biotech Co., Ltd. (Shanghai, China). The source temperature was set at 580℃, with Ion Source Gas 1 at 45 and Ion Source Gas 2 at 60. The Curtain Gas was maintained at 35. SIMCA-P (version 14.1, Umetrics, Umea, Sweden) and XCMS [[84]49] software were used to quantitatively analyze glycolysis-related metabolites. The screening criteria for the significantly differentially expressed metabolites in this study were a fold change > 2.0 or < 0.5, and p < 0.05. This screening criterion was derived from the published articles [[85]50–[86]52]. Quantitative detection of target proteins using PRM proteomics We downloaded a set of glycolysis-related genes from the GSEA database [[87]53, [88]54]. We intersected the significantly differentially expressed proteins obtained from the TMT proteomics analysis with the proteins encoded by glycolysis-related genes to obtain 11 significantly differentially expressed glycolysis-related proteins. The parallel reaction monitoring (PRM) quantitative proteomics was employed to detect the expression levels of significantly differentially expressed glycolysis-related proteins. Liquid chromatography-mass spectrometry/mass spectrometry analysis was performed using a Q-Exactive mass spectrometer (Thermo Fisher Scientific) that was coupled to an Easy nLC chromatograph system (Proxeon Biosystems, now Thermo Fisher Scientific) at Applied Protein Technology Biotech Co., Ltd. (Shanghai, China). The detection mode was positive ion mode with a scanning range from 300 to 1800 m/z. The maximum ion time was 200 ms. The automatic gain control target was 3e6. Skyline software [[89]55] was used to quantitatively analyze target peptide segments. However, the PRM proteomics analysis could not detect the expression of PGM2 and PGAM1 in the samples. Therefore, only nine differentially expressed proteins (ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PKM, TPI1, and PGM1) were further identified by the PRM proteomics. Furthermore, the expression data of the nine differentially expressed glycolysis-related proteins were extracted from the TMT proteomic protein expression profiles. Fluorescent quantitative polymerase chain reaction assay HeLa cells were seeded in a six-well plate and incubated with 0.16% DMSO and 65 μM 6-methoxyflavone for 48 h. A centrifuge column-based RNA extraction kit (Beyotime, Shanghai, China) was used to extract total RNA. The reverse transcription premixed kit (Agbio, Changsha, Hunan, China) was used to synthesize complementary DNA. Real-time fluorescent quantitative polymerase chain reaction (qPCR) analyses were conducted using the SYBR Green premixed qPCR kit (Agbio) and the qPCR system OSE-R96 (TIANGEN, Beijing, China). The internal reference was glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Sangon Biotech, Shanghai, China). The Delta-Delta Ct method was used to assess the relative expression levels of mRNA. Table [90]1 lists the qPCR primer sequences. Table 1. Sequences of primers utilized in the quantitative polymerase chain reaction Primer (human) Forward Reverse ALDOA ATGCCCTACCAATATCCAGCA GCTCCCAGTGGACTCATCTG ENO3 GGCTGGTTACCCAGACAAGG TCGTACTTCCCATTGCGATAGAA GPI CAAGGACCGCTTCAACCACTT CCAGGATGGGTGTGTTTGACC LDHA ATGGCAACTCTAAAGGATCAGC CCAACCCCAACAACTGTAATCT LDHB TGGTATGGCGTGTGCTATCAG TTGGCGGTCACAGAATAATCTTT PGK1 TGGACGTTAAAGGGAAGCGG GCTCATAAGGACTACCGACTTGG PGM1 CCAAACCGACTGAAGATCCGT CATGTTTCGATCCCCATCTCC PKM ATGTCGAAGCCCCATAGTGAA TGGGTGGTGAATCAATGTCCA TPI1 CTCATCGGCACTCTGAACG GCGAAGTCGATATAGGCAGTAGG [91]Open in a new tab Abbreviations: ALDOA fructose-bisphosphate aldolase A, ENO3 enolase 3, GPI glucose- 6-phosphate isomerase, LDHA lactate dehydrogenase A, LDHB lactate dehydrogenase B, PGK1 phosphoglycerate kinase 1, PGM1 phosphoglucomutase- 1, PKM pyruvate kinase, TPI1 triosephosphate isomerase Structural changes in glycolysis-related genes/proteins HeLa cells were treated with 6-methoxyflavone at a concentration of 65 μM, using 0.16% DMSO as a control, for a duration of 48 h. The control group and the treatment group were replicated three times. The treated cells were subjected to total RNA extraction (RNAiso Plus reagent, Takara, Kyoto, Japan) and RNA sequencing at Sangon Biotech Co., Ltd. (Shanghai, China). According to the sequence data of these RNAs, we analyzed the structural changes in glycolysis-related genes. The Illumina HiSeq platform (Illumina, San Diego, California, USA) was used for gene structure analysis. StringTie [[92]56] and GffCompare [[93]57] were used to identify new transcripts. An ASprofile [[94]58] was used for alternative splicing analysis. The parameters used for the analysis of alternative splicing were set as follows: file format: FASTA; analysis type: AS; event type: AE, IR, MIR, MSKIP, SKIP, TSS, TTS, XAE, XIR, XMIR, XMSKIP, XSKIP; the other parameters are the default values. Additionally, the treated cells were subjected to lysis, followed by the extraction and digestion of proteins. This methodology is consistent with the sample processing techniques employed in TMT proteomics. MS technology was used to perform protein structure analysis. InterProScan software and the InterPro member database [[95]59] were used to search for protein sequences and identify protein domain signatures. Subcellular localization analysis of glycolysis-related proteins MS technology was used to analyze the subcellular localization of glycolysis-related proteins. The CELLO system [[96]60] was used to identify the subcellular localization of the proteins. Molecular docking and non-covalent interaction analyses The 6-methoxyflavone structure was obtained from the PubChem database [[97]61]. The structures of ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, PKM, and TPI1 were obtained from the AlphaFold database [[98]62, [99]63]. AutoDock Vina [[100]64] and PyMOL 2.2 ([101]https://pymol.org/2/) were used to perform molecular docking of 6-methoxyflavone with the target protein receptor. The inhibition (Ki) and dissociation constants (Kd) were calculated based on the minimum binding free energies. We used the Protein–Ligand Interaction Profiler [[102]65] database to analyze non-covalent interactions between 6-methoxyflavone and target proteins. Expression levels of nine glycolysis-related genes in pan-cancer tissues We obtained the messenger RNA (mRNA) expression profile data for 33 cancer types from the Xena database [[103]66]. We used the R packages limma [[104]67], reshape2 [[105]68], and ggpubr ([106]https://CRAN.R-project.org/package=ggpubr) to analyze the differential expression of the nine glycolysis-related genes in normal and pan-cancer tissues. Analyses of clinical characteristics We downloaded mRNA expression profile data and clinical characteristic data from 300 cervical cancer tissue samples from The Cancer Genome Atlas (TCGA) ([107]https://www.cancer.gov/tcga). We divided the nine glycolysis-related genes into high- and low-expression groups using the median method. Progression-free survival (PFS) data were obtained from the Xena database [[108]66]. The R survival package ([109]https://CRAN.R-project.org/package=survival) was used to analyze the data. We downloaded mRNA expression and survival data from 46 cervical adenocarcinoma tissue samples obtained from TCGA. Immunological correlation analyses of nine glycolysis-related genes Analyses of nine glycolysis-related genes and the tumor microenvironment Stromal cells, immune cells, and tumor cells are important components of the tumor microenvironment. We downloaded mRNA expression profile data of 307 normal and cervical cancer tissue samples from TCGA. The R limma and estimate ([110]https://R-Forge.R-project.org/projects/estimate/) packages were used to score stromal and immune cells in tumor tissues. The ESTIMATE score is a comprehensive score for stromal and immune cells. We divided the nine glycolysis-related genes into high- and low-expression groups using the median method. The R limma, reshape2, and ggpubr packages were used to analyze the differences in stromal and immune cell scores between the high and low gene expression groups. Correlation between the nine glycolysis-related genes and immune cells The mRNA expression profiles were the same as those described above. The R CIBERSORT [[111]69] package was used to calculate the proportions of 22 types of immune cells in each cervical cancer tissue sample. The R limma, reshape2, ggpubr, ggExtra ([112]https://CRAN.R-project.org/package=ggExtra), and vioplot ([113]https://github.com/TomKellyGenetics/vioplot) packages were used to analyze and visualize the correlation between gene expression levels and the proportion of immune cells. Correlation between the nine glycolysis-related genes and immune checkpoints The mRNA expression profiles were the same as those described above. The R packages limma, reshape2, ggpubr, and corrplot ([114]https://github.com/taiyun/corrplot) were used to analyze and visualize the correlation between the mRNA expression levels of glycolytic and immune checkpoint genes in cervical cancer tissues. Relationship between the nine glycolysis-related genes and immunotherapy The mRNA expression profiles were the same as those described above. The R limma package was used to process the data. We submitted the processed data to the Tumor Immune Dysfunction and Exclusion (TIDE) database [[115]70] to obtain an immunotherapy efficacy evaluation file for patients with cervical cancer. We extracted immune response, immune escape, and mRNA expression data of glycolysis-related genes. The R limma, plyr ([116]https://www.jstatsoft.org/v40/i01/), and ggpubr packages were used to analyze and visualize the relationship between immune response, immune escape, and mRNA expression levels of glycolysis-related genes in patients with cervical cancer. We obtained the immunotherapy efficacy data for patients with cervical cancer from The Cancer Imaging Archive (TCIA) database [[117]71]. The R limma and ggpubr packages were used to analyze and visualize the relationship between the efficacy of immunotherapy for cervical cancer and the expression levels of glycolysis-related genes. Glycolysis stress test assay Three Seahorse XF24 cell culture microplates (Agilent Technologies) were utilized for the implantation of HeLa, SiHa, and HaCaT cells. The treatment concentrations of 6-methoxyflavone were 16.25 μM, 32.5 μM, and 65 μM. The duration of the treatment was 48 h. The glycolysis stress assay was performed using a Seahorse Bioscience XF24 Extracellular Flux Analyzer and Wave 2.6.3 software (Agilent Technologies). We continuously monitored glycolytic stress for 100 min. We recorded the extracellular acidification rate (ECAR) at 12 time points within the 100 min. Glucose was added to detect basal glycolysis levels. We added 56 μl of 100 mM glucose to each well of the microplates. Oligomycin is a glycolytic inducer, and it was added to detect maximum glycolytic capacity. We added 62 μl of 10 μM oligomycin to each well of the microplates. 2-Deoxy-glucose (2-DG) is a glycolytic inhibitor, and it was added to determine whether the production of ECAR originated from the glycolytic pathway. We added 69 μl of 500 mM 2-DG to each well of the microplates. Basic glycolysis = (Maximum rate measurement before oligomycin injection) – (Last rate measurement before glucose injection). Glycolytic capacity = (Maximum rate measurement after oligomycin injection) – (Last rate measurement before glucose injection). Glycolytic reserve = (Glycolytic capacity)—(Glycolysis) = (Maximum rate measurement after oligomycin injection)—(Maximum rate measurement before oligomycin injection). 2-DG response = (Last measurement rate before 2-DG injection) – (Last rate measurement after 2-DG injection). Glucose- 6-phosphate isomerase activity assay In the non-cellular enzyme activity assays, the preliminary experiment indicated that 6-methoxyflavone at a concentration of 65 µM did not affect the enzyme activity of glucose- 6-phosphate isomerase (GPI), L-lactate dehydrogenase (LDH), or pyruvate kinase (PKM) following a 15-min incubation period. Furthermore, due to the enzyme’s susceptibility to inactivation at 37 °C, prolonged incubation was deemed inappropriate. As a result, we opted to increase the concentration of 6-methoxyflavone. We incubated 0.6 U/μl glucose- 6-phosphate isomerase (GPI) (Baomanbio, Shanghai, China) with three concentrations of 6-methoxyflavone (0.48% DMSO, 97.5 μM, and 195 μM) at 37 ℃ for 15 min. Following this, the glucose- 6-phosphate isomerase activity assay kit (BoxBio, Beijing, China) was used to assess the activity of glucose- 6-phosphate isomerase. The optical density values were detected at 340 nm using a microspectrophotometer K5800 C (Kaiao Technology, Beijing, China). The volume and optical path length of the ultraviolet quartz cuvette are 1 ml and 10 mm, respectively. The percentage of enzyme activity = (The optical density values of 6-methoxyflavone treatment groups with different concentrations)/(The optical density value of the control group). L-lactate dehydrogenase activity assay We incubated 1.14 U/μl L-lactate dehydrogenase (LDH) (Solarbio, Beijing, China) with three concentrations of 6-methoxyflavone (0.48% DMSO, 97.5 μM, and 195 μM) at 37 ℃ for 15 min. Following this, the L-lactate dehydrogenase activity assay kit (Sangon Biotech) was used to assess the activity of L-lactate dehydrogenase. The optical density values were detected at 450 nm using a microplate reader MB580 (Huisong Heales). Pyruvate kinase activity assay We incubated 14 U/ml pyruvate kinase (PKM) (Solarbio) with three concentrations of 6-methoxyflavone (0.48% DMSO, 97.5 μM, and 195 μM) at 37 ℃ for 15 min. Following this, the pyruvate kinase activity assay kit (BoxBio) was used to assess the activity of pyruvate kinase. The optical density values were detected at 340 nm using a microspectrophotometer K5800 C (Kaiao Technology). The volume and optical path length of the ultraviolet quartz cuvette are 1 ml and 10 mm, respectively. On the other hand, due to the alteration in intracellular enzyme activity, which necessitates a complex process and additional time, we selected the intervention duration and concentration of 6-methoxyflavone for the intracellular enzyme activity assay to be 48 h and 65 µM, respectively. This decision was based on the experimental results from the CCK- 8 assay. Furthermore, in order to better observe the effects of varying concentrations of 6-methoxyflavone on enzyme activity, we added a treatment group with a concentration of 32.5 µM. Consequently, the HeLa cells were treated with three concentrations of 6-methoxyflavone (0.16% DMSO, 32.5 μM, and 65 μM) for 48 h. We collected the cells and prepared a crude enzyme solution using an ultrasonic homogenizer, Scientz-IID (Scientz, Ningbo, Zhejiang, China). Following this, the pyruvate kinase activity assay kit, the microspectrophotometer K5800 C, and the ultraviolet quartz cuvette were used to assess the activity of pyruvate kinase. Statistical analysis GraphPad Prism 8.4.3 ([118]https://www.graphpad.com/scientific-software/prism/) and R 4.3.2 ([119]https://www.r-project.org/) were used for statistical analyses. All results were analyzed using an unpaired or paired t-test, log-rank test, Wilcoxon test, chi-square test, or one-way analysis of variance. Tukey’s multiple comparisons test and Dunnett’s multiple comparisons test were conducted to address the issue of multiple comparisons. Significant differences were identified as p < 0.05. All experimental procedures were repeated with a minimum of three repetitions. For each CCK8 assay, six technical replicates and three biological replicates were implemented. Similarly, three biological replicates were executed for both the TMT and PRM proteomic analyses, with the PRM analysis utilizing the same samples as those used in the TMT analysis. Additionally, six biological replicates were carried out for both untargeted and targeted metabolomic analyses. Furthermore, three technical replicates and three biological replicates were performed for each qPCR assay. Moreover, five biological replicates were conducted for each glycolysis stress test assay. Finally, all enzyme activity assays were repeated three times. Results The study was divided into three parts. The first part elucidated the role and mechanism of 6-methoxyflavone in the negative regulation of glycolysis metabolism in HeLa cells through TMT proteomics, PRM proteomics, and untargeted and targeted metabolomics. The second part elaborated on the relationship between downregulated glycolysis-related genes and proteins after treatment with 6-methoxyflavone and survival prognosis, clinical characteristics, immune status, and immunotherapy of patients with cervical cancer. The third part validated the impact of 6-methoxyflavone on glycolysis in HeLa cells through glycolysis stress tests and enzyme activity assays. 6-Methoxyflavone inhibits proliferation in HeLa cells After treating HeLa cells with increasing concentrations of 6-methoxyflavone, we assessed cell viability using the CCK8 kit. Our findings revealed that 6-methoxyflavone inhibited the proliferation of HeLa cells. Figure [120]1A illustrates the inhibitory effect of 6-methoxyflavone on HeLa cells at 24, 48, and 72 h. The mean IC50 values were 112.33 μM, 71.44 μM, and 47.09 μM, respectively. Consequently, we chose 48 h for subsequent assays. After 48 h of treatment with 6-methoxyflavone on HeLa cells, the mean IC50 value was 71.44 µM, and the IC50 values for three biological replicates were 67.99 μM, 70.12 μM, and 76.21 μM, respectively. For subsequent experiments, we selected the minimum IC50 value in order to mitigate the risk of false-positive results that may arise from the use of excessively high drug concentrations. The minimum IC50 value was 67.99 μM. For the convenience of calculating the dose, we chose 65 μM for subsequent assays. Fig. 1. [121]Fig. 1 [122]Open in a new tab The results of the CCK8 assay and multi-omics analyses. A 6-Methoxyflavone inhibits proliferation in HeLa cells. Inhibition ratios of HeLa cells. The cells were treated with six concentrations of 6-methoxyflavone (0.16% DMSO, 10 μM, 20 μM, 40 μM, 80 μM, and 100 μM) for 24, 48, and 72 h. Six technical replicates were set for each concentration. Three biological replicates were performed for each CCK8 assay. The CCK8 kit was used to detect the inhibitory effects. The IC50 values, mean IC50 values, and standard deviation were calculated using GraphPad Prism software. B-E Volcano plots of multi-omics. HeLa cells were treated with 0.16% DMSO and 6-methoxyflavone (65 μM) for a duration of 48 h. The graphs illustrate the variation in protein or metabolite expression levels observed between the treatment group administered 6-methoxyflavone at a concentration of 65 μM and the control group treated with 0.16% DMSO. The x-axis illustrates the log2 (fold change) in the expression levels of proteins or metabolites, with positive values signifying upregulation and negative values denoting downregulation. Conversely, the y-axis depicts the -log10 (p-value), where elevated values correspond to greater statistical significance. B Volcano plot of differentially expressed proteins. Three biological replicates were performed for all experiments. Red dots indicate proteins that are significantly upregulated (p < 0.05, fold change > 1.2), while dark blue dots represent proteins that are significantly downregulated (p < 0.05, fold change < 0.84). Gray dots are used to denote proteins that exhibit no significant change. C-D Volcano plot of untargeted metabolomics. Six biological replicates were performed for all experiments. Red dots indicate metabolites that are significantly upregulated (p < 0.05, fold change > 2.0), while light blue dots represent metabolites that are significantly downregulated (p < 0.05, fold change < 0.5). Gray dots are used to denote metabolites that exhibit no significant change. C Positive ion mode of untargeted metabolomics. D Negative ion mode of untargeted metabolomics. E Volcano plot of targeted metabolomics. Six biological replicates were performed for all experiments. Red dots indicate metabolites that are significantly upregulated (p < 0.05, fold change > 2.0), while green dots represent metabolites that are significantly downregulated (p < 0.05, fold change < 0.5). Gray dots are used to denote metabolites that exhibit no significant change. Abbreviations: p, unadjusted p-value; 24 h, 24 h; 48 h, 48 h; 72 h, 72 h; IC50, half-maximal inhibitory concentration values;‾x, average; s, standard deviation; CCK8, cell counting kit 8; DMSO, dimethyl sulfoxide 6-Methoxyflavone regulates glycolysis metabolism in HeLa cells HeLa cells were treated with 0.16% DMSO and 6-methoxyflavone (65 μM) for a duration of 48 h. We performed TMT proteomic analysis to compare protein expression profiles between the treatment group administered 6-methoxyflavone at a concentration of 65 μM and the control group treated with 0.16% DMSO. We obtained expression profile data for 6,556 proteins. Significantly differentially expressed proteins were identified with a fold change > 1.2 or < 0.84, and p < 0.05. A total of 526 significantly differentially expressed proteins were identified in the study. Of these, 193 proteins were upregulated in the treatment group compared to the control group, whereas 333 proteins were downregulated (Fig. [123]1B). Notably, many of these significantly downregulated proteins were closely associated with glycolysis, such as ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, PKM, and TPI1. Furthermore, functional and pathway enrichment analyses of the differentially expressed proteins identified seven terms that were significantly associated with glycolysis (p < 0.05) (Table [124]2). Table 2. Functional and pathway enrichment analyses of TMT proteomics data Term ID Term Name Test/Ref p FDR hsa00010 Glycolysis Gluconeogenesis 13/43 1.90E- 05 2.61E- 03 GO:0061621 Canonical glycolysis 9/20 9.66E- 06 2.09E- 03 GO:0061615 Glycolytic process through fructose- 6-phosphate 9/21 1.57E- 05 2.92E- 03 GO:0061620 Glycolytic process through glucose- 6-phosphate 9/21 1.57E- 05 2.92E- 03 GO:0006096 Glycolytic process 13/74 0.01 0.13 GO:0005977 Glycogen metabolic process 7/35 0.02 0.24 GO:0005980 Glycogen catabolic process 4/16 0.03 0.32 [125]Open in a new tab Abbreviations: p unadjusted p-value, FDR false discovery rate, Test/Ref number of differentially expressed proteins/total number of proteins in the term 6-Methoxyflavone downregulates the expression of glycolysis-related metabolites 6-Methoxyflavone significantly inhibits glycolysis in HeLa cells. HeLa cells were treated with 0.16% DMSO and 65 μM 6-methoxyflavone for 48 h. We conducted both untargeted and targeted metabolomic analyses to compare the metabolite expression profiles between the treatment group receiving 6-methoxyflavone at a concentration of 65 μM and the control group treated with 0.16% DMSO. Significantly differentially expressed metabolites were identified with a fold change > 2.0 or < 0.5, and p < 0.05. From the positive ion mode of untargeted metabolomics, we obtained expression profile data for 901 metabolites, identifying a total of 311 significantly differentially expressed metabolites. Among these, 96 metabolites were upregulated in the treatment group compared to the control group, while 215 metabolites were downregulated (Fig. [126]1C). Additionally, expression profile data for 564 metabolites were obtained from the negative ion mode of untargeted metabolomics, resulting in the identification of 154 significantly differentially expressed metabolites. Of these, 36 metabolites were upregulated in the treatment group compared to the control group, whereas 118 metabolites were downregulated (Fig. [127]1D). Furthermore, targeted metabolomics yielded expression profile data for 242 metabolites, identifying 60 significantly differentially expressed metabolites. In this case, 29 metabolites were upregulated in the treatment group compared to the control group, while 31 metabolites were downregulated (Fig. [128]1E). Notably, following treatment with 6-methoxyflavone, significant downregulation was observed for D-glucose 1-phosphate, glucose 6-phosphate, D-fructose 6-phosphate, and pyruvate, all of which are closely associated with glycolytic processes (Fig. [129]2). Supplementary File 1 presents total ion chromatograms (TIC) from untargeted metabolomics, and Supplementary File 2 presents those from targeted metabolomics. Fig. 2. [130]Fig. 2 [131]Open in a new tab Glycolysis pathway and expression levels of significantly differentially expressed glycolysis-related metabolites. HeLa cells were treated with 0.16% DMSO and 65 μM 6-methoxyflavone for 48 h. A Glycolysis pathway [[132]48]. Compared to the 0.16% DMSO control group, nine downregulated glycolytic proteins were identified through TMT and PRM proteomics, while four downregulated glycolytic metabolites were detected using untargeted and targeted metabolomics in the group treated with 6-methoxyflavone. In the TMT proteomic analysis, significantly differentially expressed proteins were identified with a fold change > 1.2 or < 0.84, and p < 0.05. Note: Inline graphic : Downregulated genes/proteins; Inline graphic : Downregulated metabolites. B Expression levels of D-glucose 1-phosphate, glucose 6-phosphate, D-fructose 6-phosphate, and pyruvate. In both untargeted and targeted metabolomics, significantly differentially expressed metabolites were identified with a fold change > 2.0 or < 0.5, and p < 0.05. Abbreviations: p, unadjusted p-value; KEGG, Kyoto Encyclopedia of Genes and Genomes 6-Methoxyflavone downregulates the expression of glycolysis-related genes/proteins We identified 526 glycolysis-related proteins through TMT proteomics. We obtained 62 glycolytic genes from the GSEA database. We intersected 526 significantly differentially expressed proteins with the 62 glycolytic gene-encoded proteins, resulting in the identification of 11 significantly differentially expressed glycolysis-related proteins: ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PKM, TPI1, PGM1, phosphoglucomutase- 2 (PGM2), and phosphoglycerate mutase 1 (PGAM1). Quantitative PRM proteomics further identified nine differentially expressed proteins, which include ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PKM, TPI1, and PGM1 (Figs. [133]2A and [134]3A). We extracted expression data for these nine proteins from TMT proteomics (Fig. [135]3B). Supplementary File 3 includes chromatograms and mass spectrograms derived from TMT proteomics, while Supplementary File 4 contains total ion chromatograms obtained from PRM proteomics. Furthermore, we employed qPCR to assess the mRNA expression levels of these nine glycolytic genes. The results indicated that in the group treated with 65 µM 6-methoxyflavone, the expression level of ENO3 was significantly downregulated compared to the 0.16% DMSO control group. Additionally, the expression levels of ALDOA, GPI, LDHA, LDHB, PGK1, and PKM were also downregulated, while PGM1 and TPI1 exhibited upregulation (Fig. [136]3C). To summarize, in the qPCR assay and proteomics, although no significant differences were detected for ALDOA, GPI, LDHA, LDHB, PGK1, and PKM between the 6-methoxyflavone-treated and DMSO groups, the expression trends were all downregulated. Notably, the qPCR results for PGM1 and TPI1 were inconsistent with those obtained from TMT and PRM proteomic analyses. The qPCR assay measures gene expression at the RNA level, while TMT and PRM proteomics assess it at the protein level. The representation of RNA and protein in vivo for gene expression may exhibit inconsistencies [[137]72, [138]73] due to the influence of various post-transcriptional regulatory mechanisms [[139]74, [140]75] and post-translational modification processes [[141]76, [142]77], as observed in our study (e.g., PGM1 and TPI1). Proteins serve as the fundamental material basis of life and are the crucial undertakers and functional executors of life activities [[143]78]. Compared to RNA, protein is more stable, and the expression levels of proteins can more accurately reflect the final biological phenotypic changes [[144]79]. Therefore, in our study, although there are discrepancies between qPCR and proteomic data (e.g., PGM1 and TPI1), we ultimately focused on protein expression levels. Fig. 3. Fig. 3 [145]Open in a new tab Expression analyses and molecular docking of glycolysis-related proteins. A-B Expression levels of glycolysis-related proteins in TMT proteomics and PRM proteomics. A PRM proteomics. B TMT proteomics. C The relative mRNA expressions of ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PKM, TPI1, and PGM1. Three technical replicates and three biological replicates were performed for each qPCR assay. Statistical analysis was conducted utilizing the paired t-test. Significant differences were identified as p < 0.05. D-V. Molecular docking and non-covalent interactions between 6-methoxyflavone and target protein receptors. D-G, L-O, and T Results of molecular dockings between 6-methoxyflavone and receptors. H-K, P-S, and U Non-covalent interactions between 6-methoxyflavone and receptors. V Legend symbols in H–K, P-S, and U. Abbreviations: p, unadjusted p-value; TMT, tandem mass tag; PRM, parallel reaction monitoring; ALDOA, fructose-bisphosphate aldolase A; ENO3, enolase 3; GPI, glucose- 6-phosphate isomerase; LDHA, lactate dehydrogenase A; LDHB, lactate dehydrogenase B; PGK1, phosphoglycerate kinase 1; PGM1, phosphoglucomutase- 1; PKM, pyruvate kinase; TPI1, triosephosphate isomerase 6-Methoxyflavone alters the structure of glycolysis-related genes or proteins After treatment with 6-methoxyflavone, the alternative splicing types and number of splicing positions of ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, PKM, and TPI1 were altered in HeLa cells (Table [146]3). New transcripts for ALDOA, ENO3, GPI, LDHA, and LDHB were identified (Table [147]4). Interactions between proteins and small-molecule drugs often occur within functional domains. Changes in these domains can cause changes in protein function. After treatment with 6-methoxyflavone, the functional domains of ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, and TPI1 were altered (Table [148]5). Table 3. Alternative splicing analysis of glycolysis-related genes Group Gene AE IR MSKIP SKIP TSS TTS XAE XMSKIP XSKIP 0.16%DMSO ALDOA 12 0 4 38 22 3 0 0 10 65 μM ALDOA 4 0 2 4 8 1 0 0 0 0.16%DMSO ENO3 4 6 2 14 27 3 4 0 0 65 μM ENO3 0 0 6 2 19 3 0 0 0 0.16%DMSO GPI 4 0 2 12 12 2 0 0 0 65 μM GPI 0 0 0 0 0 0 0 0 0 0.16%DMSO LDHA 0 0 16 36 22 3 12 10 18 65 μM LDHA 0 0 14 34 23 3 8 8 14 0.16%DMSO LDHB 0 0 0 0 11 3 0 0 0 65 μM LDHB 0 0 0 0 10 3 0 0 0 0.16%DMSO PGK1 0 0 0 0 12 3 0 0 0 65 μM PGK1 0 0 0 0 12 3 0 0 0 0.16%DMSO PGM1 0 0 0 0 10 3 0 0 0 65 μM PGM1 0 0 0 0 10 3 0 0 0 0.16%DMSO PKM 0 0 0 22 36 9 0 18 34 65 μM PKM 0 0 0 16 35 9 0 18 28 0.16%DMSO TPI1 0 0 0 0 12 2 0 0 0 65 μM TPI1 0 0 0 0 18 3 0 0 0 [149]Open in a new tab Abbreviations: AE alternative exon ends, XAE approximate AE, IR intron retention, TSS transcription start site, TTS transcription terminal site, SKIP skipped exon, XSKIP approximate SKIP, MSKIP multi-exon SKIP, XMSKIP approximate MSKIP Table 4. New transcript analysis of glycolysis-related genes Gene Isoform ID Position Class Code Compare Ref ALDOA MSTRG.4913.3 16[+]30064739–30070352 j ENST00000395248 ENO3 MSTRG.5309.3 17[+]4947520–4948544 o ENST00000519266 ENO3 MSTRG.5309.2 17[+]4947512–4,948,532 o ENST00000519266 ENO3 MSTRG.5309.4 17[+]4950190–4957118 j ENST00000323997 GPI MSTRG.6721.2 19[+]34364999–34,400,276 j ENST00000588991 LDHA MSTRG.2160.1 11[+]18394560–18407847 j ENST00000422447 LDHA MSTRG.2160.4 11[+]18394704–18407847 j ENST00000422447 LDHB MSTRG.2893.1 12[-]21635342–21,737,698 j ENST00000396076 [150]Open in a new tab Table 5. Functional domain analysis of glycolysis-related proteins Protein Gene Domain Domain_Name Start_location Stop_location [151]P04075 ALDOA PF00274 Fructose-bisphosphate aldolase class-I 15 364 [152]P13929 ENO3 PF03952 Enolase, N-terminal domain 4 134 [153]P13929 ENO3 PF00113 Enolase, C-terminal triosephosphate isomerase barrel domain 143 430 [154]P06744 GPI PF00342 Phosphoglucose isomerase 54 546 [155]P00338 LDHA PF00056 lactate/malate dehydrogenase, nicotinamide adenine dinucleotide binding domain 22 160 [156]P00338 LDHA PF02866 lactate/malate dehydrogenase, alpha/beta C-terminal domain 164 324 [157]P07195 LDHB PF02866 lactate/malate dehydrogenase, alpha/beta C-terminal domain 165 330 [158]P07195 LDHB PF00056 lactate/malate dehydrogenase, nicotinamide adenine dinucleotide binding domain 23 161 [159]P00558 PGK1 PF00162 Phosphoglycerate kinase 10 406 [160]P36871 PGM1 PF02878 Phosphoglucomutase/phosphomannomutase, alpha/beta/alpha domain I 15 157 [161]P36871 PGM1 PF02879 Phosphoglucomutase/phosphomannomutase, alpha/beta/alpha domain II 194 300 [162]P36871 PGM1 PF02880 Phosphoglucomutase/phosphomannomutase, alpha/beta/alpha domain III 306 420 [163]P36871 PGM1 PF00408 Phosphoglucomutase/phosphomannomutase, C-terminal domain 464 525 [164]P60174 TPI1 PF00121 Triosephosphate isomerase 8 245 [165]Open in a new tab Subcellular localization analysis of glycolysis-related proteins Interactions between 6-methoxyflavone and ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, PKM, and TPI1 may occur in the cytoplasm of HeLa cells (Table [166]6). Table 6. Subcellular location of glycolysis-related proteins Targets Entry Peptide Sequence Organism Subcellular location ALDOA [167]P04075 GILAADESTGSIAK Homo sapiens (Human) Cytoplasm ENO3 [168]P13929 FGANAILGVSLAVCK Homo sapiens (Human) Cytoplasm GPI [169]P06744 EWFLQAAK Homo sapiens (Human) Cytoplasm LDHA [170]P00338 VTLTSEEEAR Homo sapiens (Human) Cytoplasm LDHB [171]P07195 GLTSVINQK Homo sapiens (Human) Cytoplasm PGK1 [172]P00558 ITLPVDFVTADK Homo sapiens (Human) Cytoplasm PGM1 [173]P36871 ADNFEYSDPVDGSISR Homo sapiens (Human) Cytoplasm PKM [174]P14618 DPVQEAWAEDVDLR Homo sapiens (Human) Cytoplasm TPI1 [175]P60174 HVFGESDELIGQK Homo sapiens (Human) Cytoplasm [176]Open in a new tab Molecular docking and non-covalent interaction analyses The minimum free energy levels between 6-methoxyflavone and target protein receptors were ≤ − 6.3 kcal/M (Fig. [177]3D-V; Table [178]7). The corresponding Ki/Kd values were ≤ 23.82 μM (Fig. [179]3D-V; Table [180]7). 6-Methoxyflavone had the strongest affinity for PKM (Fig. [181]3O, S; Table [182]7). The main types of non-covalent interactions between 6-methoxyflavone and nine target protein receptors were hydrophobic interaction, hydrogen bond, and π-stacking (perpendicular) (Fig. [183]3D-V; Table [184]7). Table 7. Molecular docking and non-covalent interaction parameters Targets Ligand Free energy (kcal/M) Ki/Kd (μM) Interactions ALDOA 6MF − 6.3 23.82 Hydrophobic Interaction Hydrogen Bond ENO3 6MF − 7.2 5.21 Hydrophobic Interaction Hydrogen Bond π-Stacking GPI 6MF − 8.2 0.96 Hydrophobic Interaction Hydrogen Bond LDHA 6MF − 7.3 4.4 Hydrophobic Interaction π-Stacking LDHB 6MF − 7.2 5.21 Hydrophobic Interaction Hydrogen Bond π-Stacking PGK1 6MF − 7.7 2.24 Hydrophobic Interaction Hydrogen Bond PGM1 6MF − 8.3 0.81 Hydrophobic Interaction PKM 6MF − 8.9 0.29 Hydrophobic Interaction Hydrogen Bond TPI1 6MF − 7.3 4.4 Hydrophobic Interaction Hydrogen Bond [185]Open in a new tab Abbreviations: Ki inhibition constant, Kd dissociation constant, 6MF 6-methoxyflavone Expression levels of 9 glycolysis-related genes in pan-cancer tissues Pan-cancer analysis showed that ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, PGM1, PKM, and TPI1 were significantly overexpressed in various cancer tissues compared to normal tissues (Fig. [186]4). 6-Methoxyflavone inhibited the glycolytic pathway, downregulated the expression of ALDOA, ENO3, GPI, LDHA, LDHB, PGK1, and PKM, and may be a potential therapeutic drug for cancer. Fig. 4. [187]Fig. 4 [188]Open in a new tab Differential expression analyses of nine glycolysis-related genes in 33 pan cancers. ns: no significance; ^*: p < 0.05, ^**: p < 0.01, ^***: p < 0.001. Significant differences were identified as p < 0.05. Abbreviations: p, unadjusted p-value; BLCA, bladder cancer; BRCA, breast cancer; CESC, cervical cancer; CHOL, bile duct cancer; COAD, colon cancer; ESCA, esophageal cancer; GBM, glioblastoma; HNSC, head and neck cancer; KICH, kidney chromophobe; KIRC, kidney clear cell carcinoma; KIRP, kidney papillary cell carcinoma; LIHC, liver cancer; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; PCPG, pheochromocytoma & paraganglioma; PRAD, prostate cancer; READ, rectal cancer; SARC, sarcoma; STAD, stomach cancer; THCA, thyroid cancer; UCEC, endometrioid cancer Analyses of clinical characteristics The mRNA expression levels of ALDOA, ENO3, GPI, LDHA, PGK1, and PKM significantly correlated with the histological classification of cervical cancer (p < 0.05). The mRNA expression levels of ALDOA, GPI, and PGK1 significantly correlated with the primary tumor condition and the Federation International of Gynecology and Obstetrics (FIGO) stage of cervical cancer (p < 0.05). The mRNA expression level of LDHA significantly correlated with the type of cervical cancer treatment (p < 0.05). The mRNA expression level of GPI correlated with lymph node invasion in cervical cancer (Table [189]8). The overall survival rates of patients with cervical cancer in the high- and low-expression groups of LDHA, PGK1, PKM, and PGM1 were significantly different (p < 0.05) (Fig. [190]5A-D). The overall survival rates of patients with cervical adenocarcinoma in the LDHA, PGK1, and ALDOA low-expression groups were significantly better than those in the high-expression group (p < 0.05) (Fig. [191]5E-G). The PFS of patients with cervical cancer in the PGK1 low-expression group was significantly better than that in the high-expression group (p < 0.05) (Fig. [192]5H). In summary, glycolysis-related genes play important roles in the prognosis of patients with cervical cancer. Table 8. Relationships between expression levels of glycolysis-related genes and the clinical parameters of cervical cancer patients Characteristics, n = 300 ALDOA ENO3 GPI LDHA PGK1 PKM high low p high low p high low p high low p high low p high low p Histological classification < 0.001 < 0.001 < 0.001 0.03 0.03 < 0.001  Squamous cell carcinoma 139 113 111 141 138 114 133 119 133 119 140 112  Adeno-carcinoma 11 37 39 9 12 36 17 31 17 31 10 38 Primary tumor 0.03 0.45 0.003 0.29  < 0.001 0.51  T1-T2 94 113 105 102 92 115 96 111 89 118 97 110  T3-T4 20 10 13 17 22 8 17 13 24 6 16 14 Lymph node invasion 0.07 0.91 0.05 0.42 0.67 0.15  N0 54 76 65 65 53 77 60 70 53 77 56 74  N1 33 26 29 30 33 26 31 28 26 33 32 27 FIGO stage 0.01 0.65 0.03 0.07 0.01 0.17  Stage I-II 103 125 116 112 105 123 108 120 105 123 108 120  Stage III-IV 42 23 31 34 40 25 39 26 42 23 37 28 Treatment type 0.13 0.3 0.3 0.03 0.08 0.42  Pharmaceutical Therapy, NOS 67 80 69 78 69 78 64 83 66 81 77 70  Radiation Therapy, NOS 83 70 81 72 81 72 86 67 84 69 73 80 [193]Open in a new tab The chi-square tests were used for statistical analyses Abbreviations: FIGO Federation International of Gynecology and Obstetrics Fig. 5. [194]Fig. 5 [195]Open in a new tab Survival analyses. A-D Overall survival of patients with cervical cancer; E-G Overall survival of patients with cervical adenocarcinoma; H Progression-free survival of patients with cervical cancer. Significant differences were identified as p < 0.05. Abbreviations: p, unadjusted p-value; PFS, progression-free survival Immunological correlation analyses of the nine glycolysis-related genes Analyses of the nine glycolysis-related genes and the tumor microenvironment Stromal, immune, and tumor cells are important components of the tumor microenvironment. Groups with low ENO3 and LDHB expression had higher immune and ESTIMATE scores (Fig. [196]6A, C). Groups with low GPI, PGK1, and TPI1 expression had higher stromal, immune, and ESTIMATE scores (Fig. [197]6B, D, E). 6-Methoxyflavone downregulated the expression levels of ENO3, LDHB, GPI, and PGK1 and was more likely to improve stromal, immune, and ESTIMATE scores, which may be helpful for the prognosis of patients with cervical cancer. Fig. 6. [198]Fig. 6 [199]Open in a new tab Analyses of tumor microenvironment and immunotherapy. Significant differences were identified as p < 0.05. A-E Analyses of the nine glycolysis-related genes and the tumor microenvironment. A The group with low ENO3 expression had higher immune and ESTIMATE scores. B The group with low GPI expression had higher stromal, immune, and ESTIMATE scores. C The group with low LDHB expression had higher immune and ESTIMATE scores. D-E The group with low PGK1 and TPI1 expression had higher stromal, immune, and ESTIMATE scores. ns: no significance; ^*: p < 0.05, ^**: p < 0.01, ^***: p < 0.001. F-J. Relationship between the nine glycolysis-related genes and immunotherapy. F-G Data from TIDE. H-J. Data from TCIA. F The immune escape scores of patients with cervical cancer in the LDHB low-expression group were significantly lower than those in the high-expression group. G Patients with cervical cancer in the LDHB low-expression group showed significantly better responses to immunotherapy than those in the high-expression group. H-I The immunophenotypic scores of patients with cervical cancer in the LDHA and PGK1 low-expression groups were significantly higher than those in the high-expression group. J The anti-CTLA4 immunotherapy effect of patients with cervical cancer in the PGK1 low-expression group was significantly better than that in the high-expression group. Abbreviations: p, unadjusted p-value; TME, tumor microenvironment; TIDE, Tumor Immune Dysfunction and Exclusion; TCIA, the Cancer Imaging Archive; IPS, immunophenotype score; Ctla4, cytotoxic T lymphocyte-associated antigen 4; Pd1, programmed cell death protein 1 Correlation between the nine glycolysis-related genes and immune cells The criteria for identifying significantly correlated immune cells required a correlation coefficient ≥ 0.3 and p < 0.05 with at least two genes. Three significantly correlated immune cell types were identified. We extracted the correlation coefficients between these three immune cell types and the nine glycolysis-related genes (Table [200]9). Subsequently, we presented the findings that indicated significant statistical differences. The mRNA expression levels of ALDOA, GPI, LDHA, PGK1, PGM1, PKM, and TPI1 negatively correlated with those of naive B cells (p < 0.05) (Table [201]9). 6-Methoxyflavone downregulated the expression of ALDOA, GPI, LDHA, PGK1, and PKM and was more likely to upregulate the proportion of naive B cells, which may facilitate acquired immunity. In addition, the mRNA expression levels of ALDOA, GPI, and TPI1 negatively correlated with CD4 memory resting T cells (p < 0.05) (Table [202]9). 6-Methoxyflavone downregulated the expression of ALDOA and GPI and was more likely to upregulate the proportion of CD4 memory resting T cells, which may be helpful for secondary immunity. Moreover, the mRNA expression levels of ALDOA, GPI, LDHA, PKM, and TPI1 positively correlated with mast cell activation (p < 0.05) (Table [203]9). 6-Methoxyflavone downregulated the expression levels of ALDOA, GPI, LDHA, and PKM and was more likely to downregulate the proportion of activated mast cells. Table 9. Correlation between the nine glycolysis-related genes and the three immune cells Gene B cells naive T cells CD4 memory resting Mast cells activated correlation coefficient p correlation coefficient p correlation coefficient p ALDOA − 0.35 2.32E- 05 − 0.31 0.0002 0.25 0.002 ENO3 0.21 0.01 − 0.16 0.06 0.02 0.81 GPI − 0.32 0.0001 − 0.32 9.78E- 05 0.19 0.02 LDHA − 0.35 2.2E- 05 − 0.13 0.11 0.31 0.0002 LDHB − 0.14 0.10 0.00 0.96 0.02 0.86 PGK1 − 0.21 0.01 − 0.02 0.78 0.16 0.07 PGM1 − 0.37 5.05E- 06 − 0.14 0.11 0.16 0.06 PKM − 0.41 5.45E- 07 − 0.07 0.41 0.33 4.88E- 05 TPI1 − 0.26 0.002 − 0.19 0.03 0.18 0.03 [204]Open in a new tab Correlation between the nine glycolysis-related genes and immune checkpoints We analyzed the relationships between the nine glycolysis-related genes and immune checkpoints. The expression of LDHB, PGM1, and PKM significantly and positively correlated with that of cluster of differentiation 276 (CD276) (p < 0.05, Table [205]10). The expression of LDHA and PKM significantly and positively correlated with that of cluster of differentiation 44 (CD44) (p < 0.05, Table [206]10). The expression of ALDOA and PKM significantly and positively correlated with that of cluster of differentiation 70 (CD70) (p < 0.05, Table [207]10). ALDOA expression significantly and positively correlated with that of tumor necrosis factor superfamily member 9 (TNFSF9) (p < 0.05, Table [208]10). 6-Methoxyflavone downregulated the expression levels of LDHB, PKM, LDHA, and ALDOA, which may be helpful in inhibiting immune checkpoints and contributing to its anticancer effects. Table 10. Correlation between glycolysis-related genes and immune checkpoints Glycolysis Immune checkpoint correlation coefficient P LDHB CD276 0.30 6.50E- 08 PGM1 CD276 0.33 2.10E- 09 PKM CD276 0.40 2.64E- 13 LDHA CD44 0.31 2.23E- 08 PKM CD44 0.35 1.96E- 10 PKM CD70 0.31 2.58E- 08 ALDOA CD70 0.32 6.70E- 09 ALDOA TNFSF9 0.36 1.54E- 10 [209]Open in a new tab Abbreviations: CD276 cluster of differentiation 276, CD44 cluster of differentiation 44, CD70 cluster of differentiation 70, TNFSF9 tumor necrosis factor superfamily member 9 Relationship between the nine glycolysis-related genes and immunotherapy We analyzed the relationships between the nine glycolysis-related genes and immunotherapy. Subsequently, we presented the results that demonstrated significant statistical differences. The immune escape scores of patients with cervical cancer in the LDHB low-expression group were significantly lower than those in the high-expression group (p < 0.05) (Fig. [210]6F). 6-Methoxyflavone downregulated the expression of LDHB, which may be helpful in preventing immune escape. Additionally, a better anti-tumor immune response is typically correlated with improved efficacy of immunotherapy. Patients with cervical cancer in the LDHB low-expression group showed a significantly better response to immunotherapy than those in the high-expression group (p < 0.05) (Fig. [211]6G). Furthermore, higher immunophenotypic scores (IPS) are generally associated with enhanced effectiveness of immunotherapy. The immunophenotypic scores of cervical cancer patients in the low LDHA and PGK1 expression groups were significantly higher than those in the high-expression group (p < 0.05) (Fig. [212]6H-I). The anti-cytotoxic T lymphocyte-associated antigen- 4 (CTLA4) immunotherapy effect in patients with cervical cancer in the PGK1 low-expression group was significantly better than that in the high-expression group (p < 0.05) (Fig. [213]6J). To sum up, 6-methoxyflavone downregulated the expression of LDHB, LDHA, and PGK1, which may be helpful in enhancing the effectiveness of immunotherapy and preventing immune escape. 6-Methoxyflavone regulates glycolysis metabolism Glycolytic stress was monitored continuously for 100 min (Fig. [214]7A, F, K). Compared to the control group (0.16% DMSO), the basal glycolysis levels of HeLa cells in the 6-methoxyflavone treatment groups at different concentrations were downregulated after the addition of glucose (Fig. [215]7B). Compared to the control group, the maximum glycolytic capacity of HeLa cells in the 6-methoxyflavone treatment groups at different concentrations was downregulated after adding oligomycin (Fig. [216]7C). Among them, the 65 μM group showed the most significant downregulation. Compared to the control group, the glycolytic reserves of HeLa cells in the 6-methoxyflavone treatment groups at different concentrations were significantly downregulated (Fig. [217]7D). 2-DG reversed the oligomycin-induced upregulation of glycolysis levels (Fig. [218]7A, E). In comparison to the control group, a higher concentration of 6-methoxyflavone resulted in a more pronounced inhibition of the glycolytic capacity of HeLa cells (Fig. [219]7E). Finally, compared to the control group, SiHa cells and HaCaT cells showed no significant changes in basic glycolysis, glycolytic capacity, glycolytic reserve, and responsiveness to 2-DG after treatment with different concentrations of 6-methoxyflavone (Fig. [220]7F-O). Fig. 7. [221]Fig. 7 [222]Open in a new tab Glycolysis stress test assay. The cells were treated with four concentrations of 6-methoxyflavone (0.16% DMSO, 16.25 μM, 32.5 μM, and 65 μM) for 48 h. Five biological replicates were performed for each assay. Significant differences were identified as p < 0.05. A-E Glycolysis stress test assay of HeLa cells. F-J Glycolysis stress test assay of SiHa cells. K-O Glycolysis stress test assay of HaCaT cells. A, F, K Continuous monitoring of the glycolytic stress curve for 100 min. Glucose was added to detect basal glycolysis levels. Oligomycin (oligomyci) was added to detect maximum glycolytic capacity. 2-DG was added to determine whether the production of ECAR originates from the glycolytic pathway. B Basal glycolysis levels. Compared with the control group (0.16% DMSO), the basal glycolysis levels of HeLa cells treated with different concentrations of 6-methoxyflavone decreased. C Glycolytic capacity. Compared with the control group, the maximum glycolytic capacity of HeLa cells treated with different concentrations of 6-methoxyflavone decreased. Among them, the 65 μM group showed the most significant downregulation. D Glycolytic reserve. Compared with the control group, the glycolytic reserves of HeLa cells treated with different concentrations of 6-methoxyflavone were significantly downregulated. E 2-DG reversed the upregulation of glycolysis levels by oligomycin. F-O Compared to the control group, SiHa cells and HaCaT cells showed no significant changes in basic glycolysis, glycolytic capacity, glycolytic reserve, and responsiveness to 2-DG after treatment with different concentrations of 6-methoxyflavone. All experiments were repeated five times. Statistical analysis was performed using a one-way analysis of variance (ANOVA). ns: no significance; ^*: p < 0.05; ^**: p < 0.01; ^***: p < 0.001. Abbreviations: 2-DG: 2-deoxy-glucose; ECAR, extracellular acidification rate 6-Methoxyflavone inhibits pyruvate kinase activity In the non-cellular enzyme activity assays, 6-methoxyflavone at concentrations of 97.5 µM and 195 µM did not significantly affect the enzyme activity of glucose- 6-phosphate isomerase (GPI) or L-lactate dehydrogenase (LDH) when compared to the control group following a 15-min incubation period (Fig. [223]8A-B). Conversely, the same concentrations of 6-methoxyflavone significantly downregulated the enzyme activity of pyruvate kinase (PKM) under the same conditions (Fig. [224]8C). Fig. 8. [225]Fig. 8 [226]Open in a new tab The enzyme activity assays. A-C The enzyme was incubated with three concentrations of 6-methoxyflavone (0.48% DMSO, 97.5 μM, and 195 μM) at 37 ℃ for 15 min. A Glucose- 6-phosphate isomerase activity assay. B L-lactate dehydrogenase activity assay. C Pyruvate kinase activity assay. D The HeLa cells were treated with three concentrations of 6-methoxyflavone (0.16% DMSO, 32.5 μM, and 65 μM) for 48 h. Intracellular pyruvate kinase activity assay. The percentages of enzyme activity were detected by a microplate reader or a microspectrophotometer. All experiments were repeated three times. Significant differences were identified as p < 0.05. Statistical analysis was performed using a one-way analysis of variance (ANOVA). ns: no significance; ^*: p < 0.05; ^**: p < 0.01; ^***: p < 0.001 On the other hand, in the intracellular enzyme activity assay, 6-methoxyflavone at concentrations of 32.5 µM and 65 µM significantly downregulated the enzyme activity of pyruvate kinase (PKM) when compared to the control group following a 48-h incubation period (Fig. [227]8D). The results of the non-cellular and intracellular pyruvate kinase activity assays were consistent with the findings related to pyruvate metabolite expression levels in metabolomics, as well as the expression levels of PKM at both the protein and mRNA levels, as observed in proteomics and quantitative PCR analyses. Discussion Previous studies have shown that glycolysis-related small-molecule compounds in traditional Chinese medicines have good anticancer effects. Bergenin [[228]10] and isoalantolactone [[229]80] exert anticancer effects by regulating the glycolysis pathway in cancer cells. 6-Methoxyflavone inhibits energy metabolism in HeLa cells via the glycolytic pathway. After treatment with 6-methoxyflavone, the basal glycolytic level, maximum glycolytic capacity, glycolytic reserve, and expression levels of glycolysis-related genes, proteins, and metabolites in HeLa cells were all downregulated. 6-Methoxyflavone inhibited pyruvate kinase activity. Alternative splicing, new transcript, and domain analyses showed that 6-methoxyflavone altered the structure and function of glycolysis-related genes and proteins. Subcellular localization analysis revealed that 6-methoxyflavone interacted with glycolysis-related proteins in the cytoplasm of HeLa cells. 6-Methoxyflavone inhibited glycolysis in the cytoplasm of HeLa cells. 6-Methoxyflavone had a good affinity for nine glycolysis-related proteins with non-covalent interaction sites. Glycolytic metabolism is closely associated with tumor progression. Tumor progression can be inhibited by inhibiting glycolysis. 6-Phosphofructose- 2-kinase inhibits the metastasis and malignant progression of colorectal cancer cells by inhibiting epithelial-mesenchymal transition and tumor glycolysis [[230]8]. 6-Methoxyflavone downregulated the expression of ALDOA, ENO3, GPI, LDHA, PGK1, and PKM. Clinical characteristic analyses showed that the expression of LDHA was related to lymph node invasion in cervical cancer; the expression levels of ALDOA, ENO3, GPI, LDHA, PGK1, and PKM were significantly related to the histological classification of cervical cancer; and the expression levels of ALDOA, GPI, and PGK1 were significantly related to the primary tumor and FIGO stages of cervical cancer. The level of glycolysis is related to tumor treatment. The inhibition of tumor glycolysis improves cancer radiation therapy resistance [[231]10] and cisplatin resistance [[232]80]. Bergenin inhibits glycolysis by downregulating hexokinase 2 and restores the sensitivity of radiation-resistant oral squamous carcinoma cells to radiation therapy [[233]10]. Isoalantolactone restores the sensitivity of cisplatin-resistant ovarian cancer cells to cisplatin treatment by inhibiting glycolysis [[234]80]. 6-Methoxyflavone downregulated LDHA expression. Analyses of clinical characteristics showed that LDHA expression was significantly related to the type of cervical cancer treatment. Glycolysis is closely associated with cancer prognosis. Inhibition of tumor glycolysis can improve patient prognosis. 6-Phosphofructose- 2-kinase improves patient prognosis by inhibiting epithelial-mesenchymal transition and tumor glycolysis [[235]8]. 6-Methoxyflavone downregulated the expression levels of LDHA, PKM, ALDOA, and PGK1. Survival analysis showed that the LDHA, PKM, ALDOA, and PGK1 low-expression groups had significantly better survival than the high-expression group. Our study found a close association between glycolysis levels and the immune status of cancer patients; however, it is important to note that this association was based solely on correlation analysis, rather than on conclusions derived from experimental testing. The inhibition of tumor glycolysis can enhance the efficacy of cancer immunotherapy. In T-cell immunotherapy, inhibition of glycolysis enhances the killing effect of T cells on tumor cells [[236]81]. 6-Methoxyflavone downregulated the expression of ENO3, LDHB, GPI, and PGK1 and was more likely to improve stromal, immune, and ESTIMATE scores, which may be helpful in the prognosis of patients with cervical cancer. Additionally, 6-methoxyflavone downregulated the expression of glycolytic genes closely associated with immune cells and immune checkpoints. Furthermore, 6-methoxyflavone downregulated the expression of LDHB, LDHA, and PGK1, which may enhance the efficacy of CTLA4 immunotherapy and prevent immune escape. In summary, 6-methoxyflavone may be a potential treatment option for cervical cancer. However, this study is limited to cervical cancer cells; further research should include other types of cancer cells, animal models, and clinical patient data. Conclusion 6-Methoxyflavone downregulates the expression levels of glycolysis-related genes, proteins, and metabolites, inhibits energy metabolism in HeLa cells through the glycolysis pathway, and exerts anticancer effects. 6-Methoxyflavone inhibited pyruvate kinase activity. Clinical characteristic and immunological correlation analyses showed that 6-methoxyflavone might be related to five clinicopathological parameters, prognosis, tumor microenvironment, immune cells, immune checkpoints, and immunotherapy efficacy in patients with cervical cancer. Supplementary Information [237]12885_2025_14133_MOESM1_ESM.pdf^ (472.9KB, pdf) Supplementary Material 1. Total ion chromatogram of untargeted metabolomics. [238]12885_2025_14133_MOESM2_ESM.pdf^ (690.3KB, pdf) Supplementary Material 2. Total ion chromatogram of targeted metabolomics. [239]12885_2025_14133_MOESM3_ESM.pdf^ (2.9MB, pdf) Supplementary Material 3. Chromatogram and mass spectrogram of TMT proteomics. [240]12885_2025_14133_MOESM4_ESM.pdf^ (158.6KB, pdf) Supplementary Material 4. Total ion chromatogram of PRM quantitative detection. Acknowledgements