The batch-corrected and normalized pan-cancer and normal tissue datasets were acquired from the University of California, Santa Cruz (UCSC, xena. http://xena.ucsc.edu/, accessed on 2 December 2022) datasets, which included The Cancer Genome Atlas (TCGA, encompassing 33 cancer types), Therapeutically Applicable Research to Generate Effective Treatments (TARGET, consisting of 7 pediatric cancers), and Genotype Tissue Expression (GTEx, comprising of 54 normal tissues). Kidney Chromophobe (KICH), Kidney Clear Cell Carcinoma (KIRC), Kidney Papillary Cell Carcinoma (KIRP), Glioblastoma Multiforme (GBM), Brain Lower Grade Glioma (LGG), Lung Adenocarcinoma (LUAD), and Lung Squamous Cell Carcinoma (LUSC) have a smaller number of corresponding normal samples available in the TCGA dataset. Consequently, these samples were combined with paired normal samples from the GTEx dataset. However, for other types of cancers, there were no matching normal samples found in GTEx, despite the absence of normal samples in the TCGA dataset. Specifically, expression data for cell lines was collected from the Cancer Cell Line Encyclopedia (CCLE) (45). The dataset for TCGA and GTEx was downloaded from the UCSC xena website, the dataset we downloaded was TCGA TARGET GTEx (13 datasets), the data were normalized using TPM, log2 (x+0.001) was transformed, and meta-analysis was used to make the TCGA data better compared to GTEx (https://ucsc-xena.gitbook.io/project/faq/advanced-data-and-datasets).

We retrieved single-cell sequencing datasets for bladder urothelial carcinoma (BLCA) (GSE145137), cholangiocarcinoma (CHOL) (GSE125449), head and neck squamous cell carcinoma (HNSCC) (GSE103322), kidney renal clear cell carcinoma (KIRC) (GSE171306), liver hepatocellular carcinoma (LIHC) (GSE125449), bile duct cancer (CHOL) (GSE125449), ovarian serous cystadenocarcinoma (OV) (GSE118828), prostate adenocarcinoma (PRAD) (GSE137829), stomach adenocarcinoma (STAD) (GSE183904), breast cancer (BRCA) (GSE138536), and melanoma (SKCM) GSE72056 from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). To address sample-to-sample variation in single-cell data from multiple samples, we employed the R-package “harmony” (46) for integration. This approach allowed us to effectively integrate and harmonize the datasets, enabling us to obtain more accurate and reliable results. We employed the Uniform Manifold Approximation and Projection (UMAP) function for dimensionality reduction in our visualization. This approach allowed us to effectively reduce the complexity of the data and visualize it in a more simplified and informative manner. Additionally, the Cellmarker website (47) was utilized to further enhance the accuracy of cell annotation.

The differential expression of UCK2 in cancer and normal tissues was analyzed and visualized using the “ggplot2” package in R software version 4.2.3, which is an open-source program available on The Comprehensive R Archive Network as of March 15, 2023. To assess the differential expression of UCK2 across different cancer stages, two sets of Wilcoxon tests were conducted. Meanwhile, we have compared the differential expression of cancers and adjacent tissues in TCGA datasets. Additionally, we have utilized the GPSAdb database (http://guotosky.vip:13838/gpsa/) and R programming language to investigate the expression of UCK2 in the TCGA, GTEx, and CCLE datasets.

Pan-carcinogenic tissue chips containing 20 types of tumors were purchased from Shanghai Outdo Biotech Company. Two tumor chips (non-small cell lung cancer and melanoma) were excluded because they only contained two samples. The experimental steps of immunohistochemistry mainly include: Firstly, paraffin-embedded tissue sections undergo the process of deparaffinization by heating them in an oven at 63 degrees for one hour, followed by treatment with xylene and hydration through a gradient of ethanol solutions. Secondly, the primary antibody (UCK2, 1:1500 dilution, 10511-1-AP, proteintech) was used and allowed to incubate overnight at 4°C within a controlled humidified environment. Subsequently, the corresponding secondary antibody was administered. The slides are then placed into a DAKO automatic immunohistochemistry instrument and the corresponding program is run for blocking, secondary antibody binding, and DAB color development procedures following the “Autostainer Link 48 User Guide”. Lastly, hematoxylin counterstaining and neutral resin mounting are carried out. All experimental procedures strictly adhered to the instructions provided with the kit. The staining intensity of each sample is represented by the average gray value. we performed a quantitative analysis using Fiji (1.54f) software by capturing two random microscopic images of each section. At least 3 replicates per sample. We used the average optical density (The Integrated Optical Density value divided by the area of the target distribution area) value to quantify the staining intensity. We numbered the slices according to the principle of randomization. Two experienced pathologists were invited to read the slices in a single-blind way, and summarized and analyzed according to the numbers of random groups.

The prognostic value of UCK2 was analyzed using univariate Cox regression analysis for overall survival (OS), disease-free interval (DFI), disease-specific survival (DSS), and progression-free interval (PFI). This analysis was carried out in 33 different cancer types and presented visually in a “forestplot” (version 3.1.1). To determine the optimal cutoff value for dividing UCK2 expression into two groups, we utilized the “sur_cutpoint” function from the “survminer” package. The sur_cutpoint function uses the ‘maxstat’ R package to select the maximum rank statistic, determines the best cut-off point for one or more continuous variables at a time, and finally selects the potential cut-off value with the largest difference between groups as the final cut-off value. The “survfit” and “ggsurvplot” functions were then used to analyze and visualize the survival differences through Kaplan-Meier curves.

We utilized the cBioPortal database (48) to analyze the relationship between UCK2 and copy number variation and mutation. Additionally, Pearson’s correlation analysis was conducted to determine the association between copy number variations (CNVs), DNA methylation levels, and UCK2 expression. To analyze the association between Promoter DNA methylation and UCK2, we used the UALCAN dataset (49, 50). The Tumor Mutation Burden (TMB) was determined using the “maftools” package, and the dataset on Masked Somatic Mutation from TCGA was downloaded for this purpose. Furthermore, the Homologous Recombination Deficiency (HRD) data was obtained from a study (51), while the Microsatellite Instability (MSI) data was collected from the “Bioc0ncoTK” package. Finally, this study presented the correlation between HRD, TMB, MSI, and the expression of UCK2.

The data on DNA methylation-based stemness scores (DNAss) and mRNA expression-based stemness scores (RNAss) were collected from the UCSC database to calculate their correlation with UCK2. The association between 5 mismatch repair genes (MMR) (52), 4 DNA methyltransferase genes (53), Homologous Recombination Repair (HRR)-related gene (54) and UCK2 expression in various cancers was visualized.

The gene list comprising the most relevant genes correlated with UCK2 were uploaded to the Database for Annotation, Visualization, and Integrated Discovery (DAVID, v2022q4) (https://david.ncifcrf.gov/tools.jsp), for further analysis and annotation. Then, to get enrichment results, it is recommended to select the official gene symbol in many identifiers and choose Homo sapiens in all kinds of species. Finally, the enrichment results of Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were obtained. To gain a further understanding of UCK2 biological processes, gene set enrichment analysis (GSEA) was also conducted using hallmark gene sets downloaded from the MSigDB database.

The study analyzed the impact of UCK2 on microenvironment infiltration in various types of cancer by utilizing the Estimation of Stromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) algorithm in the R package “ESTIMATE” (version 1.0.13) (55), including the assessment of ESTIMATE, stromal and immune scores. The immune checkpoint markers data collected from a previous study (56) were utilized to study the correlation of UCK2 expression, the immune subtypes data was acquired from the UCSC, and any immune subtypes containing less than three samples were excluded from the investigation into the relevance of UCK2 expression. The comparison between these subtypes was conducted in pan-cancer using the “ggstatsplot” (vision 0.11.0) package in the R programming language. TISIDB, a comprehensive repository portal dedicated to the study of tumor-immune system interactions, provides vital data on major histocompatibility complex (MHC) genes, immune stimulator and inhibitor genes, chemokine genes, and chemokine receptor genes. In the R programming language, by applying Spearman’s rank correlation test and the modified “pheatmap” package, a heatmap was generated to demonstrate the associations between these genes and UCK2 expression. The immunocyte infiltrating correlations of UCK2 were subsequently calculated using the CIBERSORT algorithm (57). The immune cell markers, obtained from a previous study (58), were explored and visualized the correlation of UCK2 expression by the packages “Hmisc”, “GSVA” and “ggplot2” in the R programming language.

The quality control of our single-cell data analysis is as follows: filter out cells with more than 20% mitochondrial gene expression, which helps to remove cells that may be abnormally expressing mitochondrial genes due to cellular stress or damage, and filter out cells with more than 2500 expression signatures, which may be to remove high-expression cells, which may represent a technical or biological abnormality, such as certain stages of the cell cycle. Cells with less than 200 expression features are filtered out, which helps remove cells with insufficient sequencing depth or high background noise. When multiple samples are involved, we use R package harmony to remove batch effects (46).

We have gathered data on the IC50 values and the corresponding gene expression of mRNA from two database - the Genomics of Drug Sensitivity in Cancer database (GDSC) and Cancer Therapeutics Response Portal (CTRP). This data has been integrated into oncopredict, a website (http://osf.io/c6tfx/) (59). Though our analysis, we have investigated the correlation between IC50 and UCK2 expression. Additionally, we have collected drug sensitivity and gene expression data from the CellMiner dataset (60) and used the R programing language to analyze the correlation between UCK2 expression and drug sensitivity.

Unless otherwise specified, all bioinformatics analysis were conducted using the R programming language. The survival disparity between the groups was illustrated using Kaplan-Meier curves. Pearson’s or Spearman’s test was utilized to determine the strength of the correlations between variables, depending on the appropriate method. When applicable, either the t-test or Wilcoxon test was utilized to analyze of the disparity between the two samples. R software (version 4.2.3 https://www.R-project.org, 15 march 2023) were used for statistical analysis and visualization. A p-value below 0.05 indicates statistical significance.

The gene UCK2 demonstrated significant expression levels in 27 type of cancer samples, it exhibits high expression levels across in 23 cancer types and low expression levels across in 4 cancer types ( Figure 1A ). We conducted an analysis of UCK2 expression using TCGA datasets with normal samples removed, as well as GTEx datasets containing normal samples, CCLE datasets containing tumor samples. In the pan-cancer analysis, UCK2 exhibited the highest level of enrichment in CESC and the lowest level of enrichment in KICH ( Figure 1B ). As presented in Figure 1C , UCK2 expression was relatively higher in various issues, including fibroblasts, lymphocytes, testes, nerves and small intestine. UCK2 may be an important target for reversing heart failure. Cardiac fibroblasts are the most important cells mediating the occurrence of myocardial fibrosis, UCK2 is highly expressed in fibroblasts, and inhibition of UCK2 expression may slow down or even reverse heart failure. Recent studies have also shown that UCK2 is an important gene related to heart failure (61). UCK2 is highly expressed in lymphocytes and may play an important role in the immune system, promoting proliferation in the T cell immune response (62) for effective response to infection or tumor. Furthermore, analysis of the CCLE dataset revealed that UCK2 was expressed in all tumor cell lines ( Figure 1D ). Moreover, the GPSAdb database (https://www.gpsadb.com/) (63) verified that UCK2 mRNA expression were consistent ( Supplementary Figures S1A–C ). The expression of UCK2 is significantly associated with the pathologic stage in five types of cancer: NHSC, KIRP, LUAD, LUSC, TGCT ( Supplementary Figures S2A–E ). In addition, we compared the expression levels of UCK2 in tumor and paired adjacent normal tissues using TCGA datasets. Our analysis revealed that UCK2 was highly expressed in several cancer types, including BLCA, BRCA, COAD, HNSC, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, and STAD, while it was found to be lowly expressed in KICH ( Supplementary Figure S1D ).

We use the experimental method of IHC to further test and verify the expression of UCK2 in various cancers. We found that the expression of UCK2 was highly expressed in cancer tissue disease compared to normal tissue, for example, BRCA, COAD, PRAD, OV, BLCA, Endometrial cancer (EC), PAAD, READ, STAD, LUAD, and Cervical squamous cell carcinoma (CSCC) ( Figure 2 ; Supplementary Figures S3 , S4 ). These findings mostly align with the results obtained from our bioinformatics analysis. Regarding the expression of UCK2 in EC and CSCC, we did not perform bioinformatics analysis. Instead, compared with cancer tissues, the UCK2 gene is highly expressed in normal tissues, including GBM, KIRC, LIHC, THCA, and Lymphadenoma ( Supplementary Figure S5 ).

To explore the prognostic value of UCK2, we analyzed the prognostic risk of UCK2 using COX regression analysis. This study examined the relationship between UCK2 mRNA expression and long-term survival across various cancer types. Our analysis revealed a strong correlation between UCK2 expression and overall survival (OS) in twelve distinct cancers, including KIRP (hazard ratio (HR) = 2.409, 95% confidence interval (CI): 1.869–3.105, p < 0.001), LIHC (HR = 1.863, 95% (CI): 1.513–2.294, p < 0.001), MESO (HR = 3.039, 95% (CI): 2.089–4.423, p < 0.001), KIRC (HR = 1.691, 95% (CI): 1.360–2.101, p < 0.001), ACC (HR = 1.995, 95% (CI): 1.411–2.820, p < 0.001), LUAD (HR = 1.241, 95% (CI): 1.109–1.389, p < 0.001), OV (HR = 0.857,95% (CI): 0.770–0.954, p = 0.005), THCA (HR = 2.426, 95% (CI): 1.276–4.611, p = 0.007), UVM (HR = 2.719, 95% (CI): 1.255–5.889, p = 0.011), PAAD (HR = 1.627, 95% (CI): 1.106–2.394, p = 0.013), KICH (HR = 2.193, 95% (CI): 1.092–4.402, p = 0.027), and SKCM (HR = 1.214, 95% (CI): 1.008–1.462, p = 0.041) ( Figure 3A ; Supplementary Figure S6 ). Additionally, the expression of UCK2 showed a significant correlation with PFI ( Figure 3B ), DSS ( Figure 3C ), and DFI ( Figure 3D ). Recently, many studies have shown that in hepatocellular carcinoma, the gene UCK2 is closely related to the immune microenvironment (43, 64), and it can promote normal and malignant T cell proliferation (62), UCK2 is closely related to LIHC survival, we believe that it may affect the tumor immune microenvironment and inhibit the activity of T cells, thus affecting the overall survival rate of patients. UCK2 is a metabolism-related gene (65), it may increase cell proliferation and inhibit cell apoptosis by promoting metabolism-related pathway activation, thereby leading to shortened survival. Studies have shown that UCK2 is related to induction of tumor cell apoptosis (66), perhaps the apoptosis of MESO and ACC cancer cells is related to the involvement of UCK2, which affects the length of survival through the apoptosis pathway.

Gene mutations can significantly impact the modulation of tumor growth and progression. We used cBioPortal (https://www.cbioportal.org/) to analyze the relationship between UCK2 expression and mutations, as well as copy-number alterations (CNAs). The mutation status of UCK2 in various cancers was evaluated and is depicted in ( Figure 4A ), we found that high UCK2 amplification in most cancers, especially cholangiocarcinoma, bladder cancer. According to Figure 4A , deep deletion is present in non-small cell lung cancer, esophageal-gastric cancer, prostate cancer (highest), and renal non-clear cell carcinoma, as indicated by the blue color. Moreover, UCK2 expression showed a positive correlation with CNV in 31 out of 33 cancers, with the exception LUAD and LAML ( Figure 4B ). We found that the gene UCK2 was highly expressed in LUAD, but it was not associated with copy number variation in LUAD. This phenomenon can be attributed to various complex biological mechanisms. One possible explanation is that even within the same cancer type, genetic backgrounds, environmental factors, and tumor heterogeneity can all influence the relationship between gene expression and copy number variation. For example, if a gene has other regulatory factors involved in its expression control, such as transcription factors or microRNAs, changes in copy number may not necessarily affect its expression levels. Furthermore, some cancers may alter gene expression through non-genetic mechanisms such as DNA methylation, histone modifications, and other epigenetic processes, which are independent of copy number variation. This could explain why there is no correlation between UCK2 expression and copy number variation in LUAD, despite the significant difference compared to normal tissue.

DNA methylation is a crucial chemical modification that plays a pivotal role in the regulation of epigenetic gene expression (67–69). In this study, our observations indicated that in 18 out of 32 types of cancers, there were statistically significant findings. Among these 18 cancers, the expression of UCK2 exhibited a negative correlation with DNA methylation levels in 16 cancers, while in 2 cancers, it showed a positive correlation. However, this correlation was found to be absent in several cancers, including DLBC, SKCM, LGG, UCEC, OV, PRAD, THYM, LUSC, ACC, COAD, THCA, HNSC, GBM, and UCS (as illustrated in Figure 4C ). Studies have shown that in LUSC, DNA hypomethylation promotes the high expression of pyrimidine metabolism rate-limiting enzymes (70), our study also found this negative correlation, but it was not statistically significant, which was contrary to our study. In another study on liver cancer prognosis, DNA methylation was negatively correlated with UCK2 expression, which obtained the same result as ours (71) Additionally, our study has revealed that the DNA methylation level of the UCK2 promoter is significantly reduced in UCEC, THCA, TGCT, PRAD, KIRP, and HNSC, while it is elevated in SARC, LUSC, KIRC, and COAD, in comparison to the adjacent normal tissues ( Figures 4D–M ). Thus, the observed abnormal increase in UCK2 mRNA expression might be linked to genetic alterations and decreased levels of DNA methylation. Moreover, in this study, we have conducted a comprehensive analysis of the correlations between TMB, MSI and HRD with UCK2 in pan-cancer. These genomic alterations are commonly observed in various types of cancers and have been shown to significantly impact patient prognosis and therapeutic responses (72, 73). As showed in Figure 4N , UCK2 demonstrated a positive correlation with TMB in 14 tumor types, namely PAAD, STAD, SARC, ACC, LUAD, PRAD, BRCA, BLCA, DLBC, READ, TGCT, LGG, HNSC, LUSC, SKCM, and KIRC and with MSI in 18 tumors (STAD, TGCT, LUSC, ESCA, LIHC, BRCA, SKCM, GBM, LGG, SARC, HNSC, UCEC, BLCA, OV, KIRP, CESC, LUAD and PRAD) ( Figure 4O ). We discovered that UCK2 was positively correlated with multiple HRD genes in most cancers, including ACC, GBM, KICH, LGG, LIHC, LUSC, MESO, PAAD, PCPG, READ, SKCM, STAD THCA, and especially UVM ( Figure 4P ). It has been reported that HRD is ubiquitous in a variety of cancers, and ATM and BRCA1/2 have been found to be important mutation drivers (74).

The maintenance of genomic stability in cancer primarily relies on the repair of DNA through various mechanisms, such as DNA MMR (52, 75) and HRR, which also play a role in preserving stemness in cancers (26, 76, 77). Therefore, we conducted an analysis of the associations between UCK2 expression and MMR-related genes, HRR signature, and tumor stemness scores (DNA stemness scores and RNA stemness scores.

Our analysis has revealed significant positive correlations between UCK2 expression and MMR-related genes in most cancers, particularly in UVM, as evidenced by ( Figure 5A ). However, we did not observe this correlation in CESC, CHOL and UCS, Interestingly, we also found positive correlations between UCK2 expression and HRR signature in most cancers, including UVM, as shown in ( Figure 5B ). It appears that UCK2 expression plays an important role in DNA damage repair.

We found a significant positive association between RNA stemness scores (RNAss) and UCK2 expression in most cancers, as shown in ( Figure 5C ). Similarly, we also found that DNA stemness scores (DNAss) were positively correlated with UCK2 expression in 9 tumors, including CESC, ESCA, HNSC, KIRC, LUAD, SARC, STAD, UCS, and especially UVM, (as depicted in Figure 5C ). Overall, this illustrates that the higher expression of UCK2 is associated with a higher tumor stemness score, stronger the activity of tumor stem cells, and lower the degree of tumor differentiation.

Various cancers show coordinated expression of DNA methyltransferases (DNMTs), as reported (79). We investigated the negative correlation between UCK2 expression and DNMTs in CESC, LIHC and PRAD ( Figure 5D ). While a positive association between UCK2 expression and DNMTs in 30 other cancers. Notably, UVM, TGCT, KICH, and GBM showed the most significant positive correlation. These results suggest that high UCK2 expression may promote DNMTs expression in diverse types of cancer.

To determine the potential function and pathway of UCK2, we conducted the GO and KEGG analysis. In BLCA, UCK2 showed significant correlated with biological processes such as cell division, DNA replication and DNA repair. Additionally, UCK2 was found to be primarily located in the nucleoplasm, nucleus and cytosol. Its most associated molecular functions were protein binding, RNA binding and ATPase activity. The signing pathway for UCK2 was identified as involving cell cycle, DNA replication and spliceosome. Cancers with similar biological functions to UCK2 in BLCA include COAD, ESCA, DLBC, HNSC, LAML, LGG, LIHC, LUAD, LUSC, MESO, READ, STAD, ACC, and THYM ( Figures 6A, B ; Supplementary Figures S7 – S13 ).

To further investigate the biological processes associated with UCK2, we conducted GSEA using hallmark gene sets. The main enriched biological functions of UCK2-related genes include DNA repair, glycolysis, E2F target, and G2M checkpoint, which are observed in several types of tumors, including ACC, BLCA, BRCA, COAD, DLBC, ESCA, HNSC, LAML, LGG, LIHC, LUAD, LUSC, MESO, READ, STAD and GBM. We also have observed that UCK2 is associated with interferon-gamma (IFNγ) response in several tumors such as ACC, BRCA, DLBC, and COAD. Previous studies have reported that IFNγ produced by T cells converts non-cancer stem cell to cancer stem cells (80). These findings suggest that UCK2 may play an essential role in maintaining genomic stability, regulating cellular metabolism, and controlling cell cycle progression in tumorigenesis, and adjusting the immune response to maintain stemness in tumor as shown in ( Figures 7A, B ; Supplementary Figures S14 – S21 ).

According to reports, UCK2 genes has been found to enhance the immune response in cases of hepatocellular carcinoma (43). Thus, we calculated the ESTIMATE of UCK2 in various types of cancer. Figure 8A illustrates that UCK2 consistently showed a negative correlation with the ESTIMATEScore, ImmuneScore, and StromScore in tumors, such as UCEC, STAD, SKCM, PRAD, OV, LUSC, LUAD, LGG, LAML, HNSC, ESCA, and BRCA. However, there was a positive correlation in KICH and KIRC. In THYM alone, UCK2 expression presented a positive correlation with the ESTIMATEScore, ImmuneScore, but a negative correlation with the StromScore. Some tumor cells can exploit immune checkpoint molecules to evade immune attacks. Thereby, increasing their chances of survival and metastasis (81). The cancers showing significant negative correlation between UCK2 expression and various immune-checkpoint-associated genes, including CHOL, DLBC, ECCA, LUSC, SKCM, TGCT, THYM, and UCS. Especially, in multiple types of cancers, a strong correlation was observed between UCK2 expression and human leucocyte antigen (HLA), mainly including HLA-I and HLA-II ( Figure 8B ). Several studies have documented the involvement of HLA molecules in promoting immune evasion by tumors (82–84). These findings suggest that the UCK2 gene may play a role in the interplay between immune checkpoints and tumor development. We Subsequently investigated whether there were differential expressions of UCK2 across various cancer immune subtypes. Our results revealed notable associations between UCK2 expression and immune subtypes in ten cancers ( Figure 8C ). It was observed that UCK2 expression was comparatively higher in the C4 immune subtype (Lymphocyte Deplete) when compared to other subtypes in several cancers, such as LUAD, STAD, SARC, and LUSC. Interestingly, there was an increase in UCK2 expression in the C2 immune subtype (IFN-gama Dominant) for BLCA and BRCA. These findings suggest that UCK2 may potentially promote tumor progression by affecting lymphocyte functions.

Furthermore, our analysis included the examination of correlations between UCK2 expression and key genes related to major histocompatibility complex (MHC), immune activation, immune suppression, chemokines, and chemokine receptors across various cancer types. As depicted in heatmaps ( Figure 8D ), UCK2 exhibited a positive correlation with genes related to five gene families in different tumor types, namely ACC, KICH, PCPG, KIRC, and UVM. This correlation was particularly significant across all five gene families, with an overall positive correlation. In contrast, it showed an overall negatively correlated with genes related to MHC in several tumors, including DLBC, ESCA, HNSC, LUSC, SKCM, TGCT, and UCS. It also exhibited negative correlations with genes related to immunoinhibitors in ESCA, LUSC, SKCM, STAD, and TGCT, with chemokine receptors in ESCA, HNSC, LUAD, LUSC, SKCM, STAD, and TGCT, with immunostimulatory genes in ESCA, LUAD, LUSC, SKCM, STAD, and TGCT, and with chemokines in ESCA, LUAD, LAML, LUSC, SKCM, and PRAD. Overall, these results suggest a close association between UCK2 expression and the biological function of immune-related genes.

To further explored the UCK2 expression in cancer immunity, we conducted a study investigating its correlation with the level of infiltration of 28 immune cell types by the ssGSEA method and Hmisc function ( Figure 9A ). our findings revealed strong positive correlations between UCK2 expression and T helper type 2 cells, activated CD4+ T cells, central memory CD8+ T cells, and memory B cells across multiple types of cancer. Conversely, we observed a significant negative correlation between UCK2 expression and T helper type 1 cells, T follicular helper cells, nature killer cells, monocytes, myeloid-derived suppressor cells (MDSCs), immature dendritic cells, immature B cells, eosinophils, effector memory CD8+ T cells, central memory CD4+ T cells, CD56bright natural killer cells, activated dendritic cells, activated CD8+ T cells, macrophages, and activated B cells within different cancer types.

Additionally, we employed the CIBERSORT algorithm to obtain 22 immunocyte correlations with UCK2. As illustrated in Figure 9B , our results indicated a positive correlation between UCK2 expression and activated memory CD4+ T cells, T follicular helper cells, M0 macrophages, M1 macrophages, and activated mast cells across several cancers. In contrast, we observed negative correlations between UCK2 expression and resting memory CD4+ T cells, monocytes, M2 macrophages, resting mast cells in various cancers. Unsurprisingly, these results were overall consistent with ssGSEA results with the aspect of T cells, monocytes, mast cells, and macrophages.

Furthermore, we conducted a comprehensive investigation into the expression of UCK2 on tumor and stromal cells in pan-cancer using the collected cancer single-cell datasets. Our findings revealed that UCK2 is significantly co-expressed on cancer cells and stromal cells in multiple cancer types, with particularly high expression observed on T cells, macrophages, cancer-associated fibroblasts, endothelial cells, cancer cells and fibroblasts ( Figure 9C ).

Our study, which utilized the CELLMINER drug response data, found that there was a positive correlation between UCK2 expression and sensitivity to drugs such as nitrogen mustard, chlorambucil, melphalan, methylprednisolone, hydroxyurea, uracil mustard, dexamethasone decadron, cladribine, fludarabine, fenretinide, nelarabine, SNS-314, sapacitabine, and CNDAC. Conversely, UCK2 expression showed a negative association with susceptibility to KU-55933 ( Figure 10A ). Furthermore, we conducted a sensitivity analysis using the GDSC (85) and CTRP databases, which revealed that high UCK2 expression in patients led to susceptibility to the top five significant drugs: ML323, BMS-345541, podophyllotoxin bromide, Eg5, and AZD5991 in GDSC2 ( Figure 10C ) drug response data, and zebularine, SB-225002, BRD-K70511574, KX2-391, and BI-2536 in CTRP2 ( Figure 10B ) drug response data. These findings suggested that dysregulation of UCK2 could lead to anti-tumor drug resistance.