RNA sequencing data and clinical details regarding UTP4 were retrieved from The Cancer Genome Atlas (TCGA) database (http://cancergenome.nih.gov/). Protein expression data were obtained from the UALCAN database (http://ualcan.path.uab.edu/index.html). Additionally, whole-genome CRISPR-Cas9 screening data were downloaded from the DepMap portal (https://depmap.org/portal/download/).

RNA sequencing and clinical data covering 33 tumor types were carefully analyzed after download from the TCGA database. Expression levels in transcripts per million (TPM) were converted using log2 transformation for accuracy. Differences in UTP4 expression across tumor types were further analyzed by constructing boxplots and paired plots in R (version 4.2.1) using packages such as ggplot2 (version 3.3.6), stats (version 4.2.1), and car (version 3.1-0). The Tumor Immune Estimation Resource version 2 (TIMER2.0; http://timer.cistrome.org/) was also used to compare UTP4 expression between tumor and normal tissues.

The UALCAN platform (http://ualcan.path.uab.edu/index.html) was utilized for additional analyses. The proteomics module within UALCAN allowed comparisons of UTP4 protein expression between tumor and normal tissues, as well as analyses of its association with clinical characteristics, including cancer stage.

To further validate differences in UTP4 expression in cancer and examine its gene expression profiles in specific cell types, the Sparkle database (https://grswsci.top/) was employed. The Spatial Transcriptomics module was selected, and five datasets were analyzed, namely ‘CESC: Human_Cervical_Cancer (10x database)’, ‘GIST2: GSE203612-GSM6177609’, ‘HNSC2: GSE181300-GSM5494476_HNSCC201125T05’, ‘KIRC2: GSE175540-GSM5924030_ffpe_c_2’, and ‘CRC2: IntestineCancer_10x_FFPE (10x database)’, to explore differences in UTP4 expression at single-cell resolution within spatial transcriptomic contexts.

RNA sequencing data and clinical information for the 33 tumor types were standardized and analyzed for prognostic significance. Using the R ‘survival’ package, Kaplan-Meier survival curves were constructed, proportional hazards assumption tests were performed, and survival regression models were fitted.

Receiver operating characteristic (ROC) curve analysis was conducted using the timeROC package (version 0.4) to evaluate diagnostic performance. The area under the ROC curve (AUC) values, ranging from 0.5 to 1, were assessed to determine diagnostic capability, with higher AUC values indicating superior diagnostic accuracy (16).

To investigate UTP4 gene alterations across different cancers, the cBioPortal platform (https://www.cbioportal.org/) was utilized, an interactive tool designed for analyzing multidimensional cancer genomic datasets. The study cohort was selected from the ‘TCGA PanCancer Atlas Studies’, and the UTP4 gene was queried using cBioPortal's module to obtain relevant mutation information. The ‘Cancer Types Summary’ module provided an overview of mutation frequencies and types associated with UTP4 across various cancers. Additionally, the ‘Mutations’ module offered detailed insights into mutation sites, counts, and types within the UTP4 gene. Mutation data were subsequently visualized using cBioPortal's visualization tools. To further analyze the mutational landscape, the ‘Mutation Effect-Expression’ module of the Sparkle database (https://grswsci.top/) was employed. This database enabled analysis of UTP4 mutation types and single nucleotide variations (SNVs), revealing potential roles for UTP4 gene alterations in cancer progression.

Multiple databases were used to investigate the relationship between UTP4 and immune processes. Initially, immune infiltration associated with UTP4 expression across various tumors was analyzed using the Xiantao Academy (www.xiantaozi.com). Subsequently, the Sparkle database (https://grswsci.top/) was utilized to examine associations between UTP4 expression and the tumor immune microenvironment, including immune cells specifically in GC. Finally, correlations between pan-cancer UTP4 expression and tumor stemness, as well as immune checkpoints, were explored using the SangerBox tool (http://sangerbox.com/home.html).

The protein interaction network involving UTP4 was obtained from the STRING database (https://string-db.org/) (17). Next, the top 100 genes correlated with UTP4 expression were retrieved from GEPIA2.0 (http://timer.cistrome.org/) (18). Utilizing the ‘clusterProfiler (version 4.4.4)’ package in R, enrichment analysis was performed on the top 10 correlated genes. This allowed the identification of crucial phenotypes and signaling pathways involving UTP4 across pan-cancer contexts. Additionally, Pearson correlations between UTP4 expression z-scores and GSVA scores for 14 tumor states were analyzed using the ‘Pathway Analysis-Activity Score’ template provided by the Sparkle database (https://grswsci.top/) (19).

The proportional hazards assumption was verified using the ‘survival’ package in R, followed by univariate Cox regression analysis conducted across various tumors. Specifically focusing on GC, multivariate Cox proportional hazards regression analysis was performed to assess the independent prognostic significance of UTP4, adjusting for clinical variables including age, sex and tumor stage. The results were illustrated through forest plots displaying hazard ratios (HRs), 95% confidence intervals (CIs), and P-values for UTP4 and other clinical factors.

Mendelian colocalization analysis, a statistical method used to investigate genetic associations with diseases across multiple phenotypes, was employed to identify shared genetic factors. To explore genetic associations of UTP4 in pan-cancer and GC contexts, datasets for pan-cancer (ieu-b-4966) and GC (ebi-a-GCST90018629) were downloaded from the OpenGWAS database. Colocalization analysis was conducted using the ‘coloc’ package. Results were visualized through the ‘locuscompare’ function within the ‘locuscompare’ package and the ‘stack_assoc_plot’ function from the ‘gassocplot2’ package. A posterior probability (PP.H4.abf) greater than 80% was set as the threshold indicating robust evidence of colocalization (20).

Genome-wide CRISPR-Cas9 screening data were obtained from the DepMap portal (https://depmap.org/portal/download/). This dataset includes dependency scores for candidate genes across numerous cancer cell lines, calculated using the CERES algorithm. The CERES dependency scores for UTP4, which were scaled such that 0 indicates a non-essential gene and -1 denotes the median dependency of core essential genes, were specifically analyzed. The 200 cell lines exhibiting the most negative UTP4 dependency scores were selected and visualized using bar plots.

The human normal gastric epithelial cell line GES-1 and GC cell lines AGS and MKN-45 were purchased from Sai Bai Kang (Shanghai) Biotechnology Co., Ltd. Cells were cultured in medium containing 5% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) at 37˚C with 5% CO₂. Cells in optimal growth conditions were transfected with CRISPR-Cas9 small guide RNA (sgRNA) lentiviral vectors. A total of three distinct sgRNAs were designed against the human UTP4 gene. Their RNA guide sequences (5'-3') are as follows: sgRNA#1: AUGCCUUUGGAGGACCUAUU; sgRNA#2: CGUCAGCAUUAGCGAUGAGA; and sgRNA#3: GUAGCCCGAGACUCAUGACC.

The knockdown efficiency of all three sgRNAs was initially evaluated by reverse transcription-quantitative PCR (RT-qPCR). sgRNA#2 showed the strongest and most consistent reduction in UTP4 mRNA levels and was therefore chosen to generate stable knockdown cell pools for all subsequent functional experiments. A custom non-targeting control (NTC) sgRNA was included in parallel to account for potential non-specific effects associated with lentiviral transduction and Cas9 activity. Its RNA guide sequence is: GGCAGCGAGAUAACUUGACUC. All sgRNA lentiviral vectors were supplied by Shenggong Bioengineering (Shanghai) Co., Ltd. The initial qPCR screening data assessing the knockdown efficiency of the three sgRNAs are provided in Fig. S1 and Table SIII.

Cell viability following sgRNA transfection was measured using the CCK-8 from Shanghai Taosu Biotechnology Co., Ltd. Transfected cells were seeded into 96-well plates at a density of 1x10³ cells per well and incubated overnight at 37˚C. At 0, 1, 2, and 3 days, 10 µl of CCK-8 solution was added to each well, and absorbance was measured at a wavelength of 450 nm.

Total RNA was extracted from cells using TRIzol Universal RNA extraction reagent according to the manufacturer's protocol. RT-qPCR were performed using the Vazyme genomic RT-qPCR kit and an ABI 7900HT Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions for PCR amplification were as follows: Initial denaturation at 95˚C for 30 sec; 40 cycles of denaturation at 95˚C for 10 sec, annealing at 60˚C for 30 sec, and extension at 72˚C for 30 sec. Quantification was performed using the 2^-ΔΔCq method (21). Each experiment was repeated three times. The primer sequences used were as follows: UTP4 forward, 5'-AGGTCCATCGAGTACGTTTCT-3' and reverse, 5'-CACAGTGCCATCTGTTCGTG-3'. GAPDH was used as an endogenous control for normalization. The primer sequences for GAPDH were as follows: forward, 5'-GAAGGTGAAGGTCGGAGTC-3' and reverse, 5'-GAAGATGGTGATGGGATTTC-3'.

Prior to the scratch assay, cells were serum-starved in medium containing 0.5% FBS for 12 h to suppress cell proliferation. A baseline was drawn horizontally at the bottom of each well in a six-well plate for reference. Once cells reached near-confluence, a sterile pipette tip was gently used to scratch a straight line along the baseline, creating a wound. Images of the initial wound state were immediately captured under a light microscope, and its width was recorded. After 48 h, images of the wound area were captured again. The wound area was precisely measured using ImageJ software (version 1.53t; National Institutes of Health) to assess cell migration rates and wound closure efficiency.

Cells were cultured in six-well plates to an appropriate density. Cell apoptosis was assessed following the manufacturer's protocol using the Annexin V-AF488/PI Apoptosis Detection Kit (BioLegend-Shanghai Yaji Biotechnology Co., Ltd.). The number of apoptotic cells was quantitatively determined using a BD LSRFortessa flow cytometer (BD Biosciences) and analyzed with BD FACSDiva software (version 9.0; BD Biosciences).

Statistical analyses were performed using R (version 4.3.0) and GraphPad Prism (version 10.3.0), with visualization and specific tests conducted via R packages (ggplot2, survminer, pheatmap, timeROC). A two-sided P<0.05 was considered to indicate a statistically significant difference.

For public omics and clinical data, continuous variables are presented as the mean ± SD or median (IQR). Group comparisons used paired/independent t-tests, one-way ANOVA (with Tukey's post hoc test), or Kaplan-Meier analysis (log-rank test). The independent prognostic value of UTP4 was assessed by Cox proportional hazards regression (assumption verified via Schoenfeld residuals). Diagnostic performance was evaluated with time-dependent ROC curves. Correlations were calculated using Pearson's coefficient. Functional enrichment (GO/KEGG) was analyzed with the clusterProfiler package (FDR-adjusted).

For in vitro data, results from ≥3 independent replicates are shown as mean ± SD. Comparisons used the independent t-test (two groups) or two-way ANOVA with Tukey's post hoc tests, as indicated.

To examine differential UTP4 expression across cancer types, RNA sequencing data from the TCGA database were analyzed. UTP4 was significantly upregulated in several cancers, including cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal cancer (ESCA), head and neck cancer (HNSC), hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), rectal adenocarcinoma (READ), stomach adenocarcinoma (STAD), bladder cancer (BLCA) and glioblastoma (GBM) (P<0.05) (Fig. 1A). These findings suggest that UTP4 may play an important role in the initiation and progression of multiple malignancies. Conversely, UTP4 expression was significantly decreased in breast cancer (BRCA), kidney chromophobe (KICH) and prostate cancer (PRAD) (P<0.05; Fig. 1A). A follow-up analysis using cancer-paired samples further supported these results. Pronounced differences in UTP4 expression were observed across the corresponding tumor types, consistent with the initial findings (Fig. 1B). Additional validation using the TIMER2.0 database confirmed similar expression patterns. Although the difference in UTP4 expression in GBM did not reach statistical significance, other cancer types exhibited expression trends consistent with the TCGA analysis. Notably, cervical cancer (CESC) showed significant upregulation of UTP4 (P<0.05; Fig. 1C).

To further characterize UTP4 expression at the protein level, proteomic data from the UALCAN database were analyzed. UTP4 protein levels were significantly elevated in multiple cancer types, including COAD, HNSC, kidney renal clear cell carcinoma (KIRC), pancreatic adenocarcinoma (PAAD), endometrial carcinoma (UCEC), LUAD, LUSC, GBM and LIHC (P<0.05; Fig. 2A). These findings reinforce the hypothesis that UTP4 may contribute to tumor progression. Moreover, UTP4 protein expression varied significantly across pathological stages in COAD, HNSC, KIRC, PAAD, UCEC and lung cancer (P<0.05; Fig. 2B). This stage-dependent variation suggests that UTP4 overexpression is closely linked to tumor development rather than being an isolated molecular event, supporting its potential role in cancer progression.

Single-cell spatial transcriptomic analysis was conducted to precisely evaluate UTP4 expression differences among various cancers and cell types (22). This approach allowed detailed exploration of UTP4 expression in specific cellular populations (23,24). The results demonstrated that in human cervical squamous cell carcinoma samples (Human_Cervical_Cancer), UTP4 expression was mainly detected in tumor and epithelial cells (Fig. 3A). Similarly, gastric stromal tumor (GSE203612-GSM6177609) samples showed predominant UTP4 expression in tumor and epithelial cells (Fig. 3B). In head and neck squamous cell carcinoma (GSE181300-GSM5494476_HNSCC201125T05), UTP4 expression was concentrated predominantly in tumor cells (Fig. 3C). In clear cell renal cell carcinoma (GSE175540-GSM5924030_ffpe_c_2) samples, UTP4 was expressed not only in tumor cells but also in macrophages (Fig. 3D). In CRC (IntestineCancer_10x_FFPE) samples, UTP4 expression was primarily localized to tumor and epithelial cells (Fig. 3E). These single-cell spatial transcriptomics findings highlight that UTP4 is predominantly expressed in tumor and epithelial cells in various tumor tissues, suggesting its critical involvement across multiple cancer types.

A comprehensive survival analysis was conducted to evaluate the prognostic significance of UTP4. Elevated UTP4 expression significantly correlated with poor prognosis in multiple cancers, including ACC (P=0.021), HNSC (P=0.008), GBMLGG (P<0.001), LUAD (P=0.010), LGG (P=0.015), LIHC (P=0.036), PAAD (P=0.025), SARC (P=0.048), STAD (P=0.023) and OSCC (P=0.038) (Fig. 4A). These results imply that UTP4 may be a robust indicator of adverse outcomes, underscoring its importance in cancer progression and patient survival. By contrast, high UTP4 expression correlated with favorable prognosis in KIRC (P<0.001) (Fig. 4A). Furthermore, ROC curve analysis evaluated the diagnostic accuracy of UTP4. UTP4 demonstrated strong diagnostic performance in KICH (AUC=0.959), STAD (AUC=0.894), LIHC (AUC=0.823) and OSCC (AUC=0.810) (Fig. 4B), and moderate diagnostic ability in HNSC (AUC=0.787), GBMLGG (AUC=0.630), LUAD (AUC=0.795), and PAAD (AUC=0.682) (Fig. 4B). Analysis of survival rates by sex showed that the association between UTP4 expression levels and prognosis was more significant in females (Fig. 4C). Notably, although UTP4 expression was significantly associated with overall survival (OS), it showed no statistically significant correlation with the disease-free interval (DFI) (P=0.39; Fig. S2). This discrepancy suggests that high expression of UTP4 may not directly drive early tumor recurrence but instead predominantly influences disease progression after recurrence. These findings emphasize the significant prognostic and diagnostic relevance of UTP4 across various cancer types. Collectively, the results support the utility of UTP4 as a potential biomarker and provide a basis for developing cancer diagnostic and prognostic assessment tools.

To improve understanding of the mutational characteristics of UTP4 across different cancers, detailed analyses were conducted. Significant variations in UTP4 mutation frequencies and types were observed among cancer types. High mutation frequencies occurred in UCEC (mutation 4.35%, amplification 0.19%, deep deletion 0.95%), BLCA (mutation 0.97%, amplification 1.95%, deep deletion 0.73%, multiple alterations 0.24%), ESAD (mutation 1.65%, amplification 1.65%, deep deletion 0.55%), PRAD (amplification 0.4%, deep deletion 2.63%), CHOL (amplification 2.78%), COADREAD (mutation 2.69%), OV (amplification 0.34%, deep deletion 2.23%), SKCM (mutation 2.03%, amplification 0.23%), BRCA (mutation 0.55%, amplification 0.18%, deep deletion 1.38%), DLBC (deep deletion 2.08%), and STAD (mutation 0.91%, amplification 0.23%, deep deletion 0.68%) (Table SI, Fig. 5A). Further analysis using cBioPortal identified R328Q and P593L/A as the most frequent mutation sites in UTP4. These mutations appeared in multiple cancers and were each supported by at least three samples (Fig. 5B). Statistical evaluation revealed that missense mutations predominated in UTP4, followed by nonsense mutations, while SNVs were mainly single nucleotide polymorphisms (SNPs) (25,26) (Fig. 5C). These findings indicate that UTP4 mutations vary significantly across cancers, suggesting diverse functional roles. Different mutation types might influence UTP4 functionality, potentially affecting cancer initiation and progression. For instance, missense mutations may alter UTP4 protein structure and function, disrupting RB and cancer cell proliferation; nonsense mutations may result in premature termination of UTP4 proteins and loss of function; and SNV-type SNPs may affect the regulation of UTP4 gene expression.

Associations between UTP4 expression, immune cells and microorganisms within the TME were further explored. UTP4 expression negatively correlated with most immune cells across pan-cancer analyses, indicating its potential role in inhibiting immune cell infiltration or function. However, positive correlations were noted with specific immune cell subsets, such as T helper cells, central memory T cells and Th2 cells (Fig. 6A). Moreover, UTP4 expression exhibited significant correlations with various microorganisms in the TME. Specifically, in STAD and COAD, UTP4 positively correlated with most microorganisms; in ESCA, it showed negative correlations; in READ and HNSC, both positive and negative correlations were observed (Fig. 6B). These results reflect the complex functional roles of UTP4 across cancers. In GC specifically, elevated UTP4 expression negatively correlated with infiltration of most immune cells but positively correlated with particular immune subsets, including Common lymphoid progenitor_XCELL, T cells CD4+ Th2_XCELL, and T cells CD4+ Th1_XCELL (Fig. 6C). Thus, UTP4 may regulate immune responses and microbial composition in the TME, influencing cancer progression and patient prognosis.

Tumor stemness refers to the stem cell-like capacity of tumor cells for self-renewal and differentiation, enabling aggressive proliferation and metastasis (27,28). Tumor stem cells significantly contribute to tumor recurrence and therapy resistance (29). To investigate the role of UTP4 in tumor stemness, correlations between UTP4 expression and tumor mutational burden (TMB) and microsatellite instability (MSI) were analyzed. UTP4 expression positively correlated with TMB in KIRC, UVM, BRCA, HNSC, LUSC, MESO, LGG, CESC, STES, PAAD, OV, READ, GBMLGG, LIHC, GBM, PRAD, DLBC, KIPAN, ACC, SARC, LUAD, STAD, UCS, KIRP and UCEC (Fig. 7A). Similarly, UTP4 expression positively correlated with MSI in LUSC, OV, BRCA, KIRP, BLCA, STES, KIRC, CESC, PAAD, UCS, GBM, LIHC, STAD, SARC, SKCM, UCEC, UVM and CHOL (Fig. 7B). Immune checkpoint molecules are crucial for tumor immune evasion. High UTP4 expression was significantly correlated with immune checkpoint molecules in multiple cancers, including DLBC, OV, KICH, PAAD, BLCA, BRCA, PCPG, PRAD, UCEC, UVM, LIHC, KIPAN, KIRC, THCA, THYM, HNSC, KIRP, LUSC and GBMLGG (Fig. 7C). These results indicate that UTP4 may influence cancer progression and metastasis by regulating tumor stemness, mutation burden and immune checkpoint expression.

To clarify the molecular mechanisms of UTP4 in cancer, enrichment analysis of UTP4-associated genes was performed. Firstly, a protein-protein interaction network involving 11 proteins related to UTP4 was obtained from the STRING database (Fig. 8A). Subsequently, using GEPIA2.0, the top 100 genes co-expressed with UTP4 were identified (Table SII). Among these, the top 10 genes, including NIP7, CIAPIN1 and SF3B3, exhibited strong pan-cancer correlations with UTP4, illustrated by scatter plots (Fig. 8B). GO and KEGG enrichment analyses (Fig. 8C and D) revealed significant involvement of UTP4 in pathways such as RB, RNA processing, and ribonucleoprotein particle assembly, highlighting its essential role in cellular functions. RB is critical for cellular protein synthesis and normal cellular operations (30,31). UTP4's participation suggests roles in various stages of ribosome assembly, potentially supporting rapid cancer cell proliferation. GO analysis results included biological processes (BP), cellular components (CC), and molecular functions (MF). The BP and CC analyses indicated that UTP4 is involved in pre-ribosome assembly and biogenesis, processes essential for efficient protein synthesis in cancer cells. Additionally, CC analysis suggested that UTP4 might participate in the formation and regulation of ribonucleoprotein granules, which influence mRNA storage and translation. Dysregulated ribonucleoprotein granules could disrupt gene expression control, thereby promoting cancer development (32). BP results also indicated that UTP4 might exhibit catalytic functions, particularly ATP-dependent RNA-related activities. Abnormal RNA processing and modification due to altered UTP4 activity may disrupt RNA metabolism, consequently supporting cancer cell proliferation and survival.

In GC, Pearson correlation analysis between UTP4 expression z-scores and GSVA scores of 14 tumor states showed significant positive correlations with the cell cycle, DNA damage and DNA repair (P<0.05; Fig. 8E). This suggests that UTP4 may facilitate cell proliferation and tumor progression, and its elevated expression might indicate increased DNA damage coupled with active repair mechanisms in tumor cells. Conversely, UTP4 expression negatively correlated with angiogenesis, differentiation, epithelial-mesenchymal transition, hypoxia, inflammation, quiescence and stemness. However, the precise roles of UTP4 in these states require further investigation.

To assess the prognostic significance of UTP4, pan-cancer univariate and GC multivariate Cox regression analyses were performed using RNA-seq data and clinical information from 33 cancer types (TCGA database). Univariate analysis revealed that UTP4 significantly influenced patient prognosis in ACC (P=0.0211), HNSC (P=0.0082), KIRC (P=2.17x10-5), LGG (P=0.0149), LIHC (P=0.0359), LUAD (P=0.0101), PAAD (P=0.0253), SARC (P=0.0485) and STAD (P=0.0226) (Fig. 9A). These findings indicate the potential utility of UTP4 as a prognostic biomarker across multiple cancers.

Subsequently, multivariate Cox regression was performed for GC (STAD), including clinical variables (pathological T stage, pathological N stage, age) with P-values <0.1 from univariate analyses along with UTP4 expression. The results demonstrated that UTP4 expression remained statistically significant (P=0.041) after adjusting for clinical factors (Fig. 9B), further validating UTP4 as an independent prognostic factor in GC.

Colocalization analysis evaluates genetic associations between genes and disease-related SNPs to elucidate potential disease mechanisms (33). Colocalization of UTP4 with pan-cancer GWAS data identified the locus rs74886619, displaying a posterior probability of 1 (Fig. 10A). This result indicates potential shared genetic mechanisms between UTP4 and pan-cancer, emphasizing its importance. Similarly, in GC (STAD), colocalization analysis identified rs113952386 with a posterior probability of 0.87 (Fig. 10B), suggesting a significant genetic role for UTP4 in GC development.

To predict the effect of UTP4 knockout on cancer cell proliferation, CRISPR-Cas9 screening data from the DepMap database were analyzed to measure gene essentiality (34). Data scaling ensured accurate scoring, with 0 representing non-essential genes and -1 indicating median effects of core essential genes. UTP4 dependency scores were frequently below -1.5, approaching -2 in various cancer cell lines, such as BLCA, BRCA and CESC (Fig. 11). These scores indicate that UTP4 is essential in these cell lines, suggesting that its knockout might significantly inhibit cancer cell proliferation or induce apoptosis, thereby confirming its role in cancer cell proliferation and survival.

Based on genetic associations from colocalization analysis and CRISPR-Cas9 screening predictions, the function of UTP4 in GC cells was further investigated. First, qPCR was employed to evaluate UTP4 mRNA expression in normal gastric epithelial cells (GES-1) and GC cell lines (AGS and MKN-45). UTP4 expression was significantly higher in GC cell lines compared with normal epithelial cells (Fig. 12A). Specific sgRNAs were designed to silence UTP4 expression effectively in both GC cell lines, and the efficacy of the selected sgRNA (sgUTP4#2) was confirmed via qPCR (Fig. 12B and C). Subsequent CCK-8, wound healing and apoptosis assays showed a marked decrease in proliferation of AGS and MKN-45 cells within 72 hours following UTP4 knockout (Fig. 12D and E). Furthermore, cell migration was significantly reduced after 48 h (Fig. 12F and G). These findings substantiate the critical role of UTP4 in GC cell migration, suggesting its involvement in invasive processes related to cancer progression. Additionally, apoptosis assays revealed a significant increase in apoptosis rates following UTP4 depletion. The apoptosis rate in AGS cells increased notably from 6.93 to 13.4%, and in MKN-45 cells from 16.04 to 28.3% (Fig. 12H and I). These results indicated that UTP4 knockout significantly inhibits GC cell proliferation and migration and promotes apoptosis, validating its critical role in GC biology.