F13B regulates angiogenesis and tumor progression in hepatocellular carcinoma via the HIF-1α/VEGF pathway

Hepatocellular carcinoma (HCC) is a highly aggressive malignant tumor with a poor prognosis. This research aimed to investigate the role of F13B in HCC and its underlying mechanisms. Through comprehensive bioinformatics analysis of the GSE120123 and The Cancer Genome Atlas (TCGA)-Liver hepatocellular carcinoma (LIHC) datasets, we identified 220 overlapping prognosis-related genes. Eight key genes, including the previously unreported CCDC170 and F13B in HCC, were identified through Least Absolute Shrinkage and Selection Operator (LASSO)-Cox regression analysis. F13B emerged as a significant prognostic factor in HCC, warranting further investigation in subsequent analyses. In vitro experiments showed that F13B expression was notably reduced in HCC cell lines and tissues, particularly in Huh-7 and SMMC-7721 cells. Overexpression of F13B inhibited cell invasion, migration, and proliferation, while its knockdown produced the opposite effect. A lactate dehydrogenase (LDH) activity assay in human umbilical vein endothelial cells (HUVECs) demonstrated that F13B overexpression reduced vascular endothelial growth factor (VEGF)-induced cytotoxicity, whereas knockdown increased it. Further analysis revealed that F13B negatively regulates VEGFA expression, affecting HUVEC proliferation. In HUVECs, F13B overexpression reversed VEGF-induced upregulation of key angiogenesis markers, including phospho-VEGF receptor 2 (p-VEGFR2), matrix metalloproteinase-2 (MMP-2), matrix metalloproteinase-9 (MMP-9), as well as AKT/mTOR signaling proteins, phospho-Akt (p-AKT), and phospho-mTOR (p-mTOR). Additionally, F13B negatively regulated VEGFA and hypoxia-inducible factor 1 A ( HIF1A ) under hypoxic conditions, counteracting the hypoxia-induced increase in cell viability. These findings suggest that F13B regulates angiogenesis through the HIF-1α/VEGF pathway and plays a crucial role in HCC progression. Our results highlight the potential of F13B as a therapeutic target in HCC, providing novel insights into the molecular mechanisms of HCC and its prognostic significance


INTRODUCTION
Hepatocellular carcinoma (HCC) is an extremely aggressive cancer characterized by irreversible changes in the liver, often triggered by factors such as drug-induced hepatitis, viral hepatitis B, idiopathic hepatitis, and local tumors (1,2).The progression of HCC to advanced stages is frequently accompanied by severe complications, including hepatic encephalopathy, upper gastrointestinal bleeding, massive ascites, rupture and bleeding of the tumor, metastatic pleural effusion, infections, fever, liver and kidney failure, and respiratory failure (3,4).These complications contribute significantly to the morbidity and mortality associated with HCC, posing a substantial threat to public health and placing a heavy burden on affected families (5).
Moreover, the treatment of HCC faces many challenges, including resistance to chemotherapy and radiotherapy, as well as a lack of effective early diagnostic and prognostic markers.In recent years, with the development of omics technologies, researchers have begun to identify molecular pathways associated with the development of HCC and potential therapeutic targets.
However, there are still many unresolved issues in the field of HCC research, including the molecular mechanisms of disease progression, the role of the tumor microenvironment, and how to more effectively translate basic research findings into clinical applications.Therefore, there is an urgent need for further research to elucidate the latent molecular pathways involved in the development and progression of HCC, in order to develop more effective prevention, diagnosis, and treatment strategies.
Vascular endothelial growth factor A (VEGFA) is a protein crucial for angiogenesis, promoting blood vessel formation and endothelial cell proliferation (6).In HCC, VEGFA is essential to the development and spread of tumors by promoting angiogenesis, facilitating tumor vascularization, and enhancing tumor cell survival (7).Zucman-Rossi et al. demonstrated that amplification of VEGFA on chromosome 6p21 is associated with HCC progression, and they also identified other genetic alterations such as FGF19/CNND1 amplification (8).These findings suggest potential therapeutic targets for advanced HCC and advocate for precision medicine approaches.Furthermore, Zhao et al. demonstrated that downregulation of microRNA-205 in HCC inhibits cell growth and metastasis by directly targeting VEGFA, offering promising therapeutic prospects (9).Additionally, Zhang et al. found that VEGFA shows increased expression in HCC tissues and serum, correlating with α-fetoprotein levels, thereby enhancing its diagnostic potential alongside KAI1 (10).The combined detection of VEGFA and KAI1 proposes a robust diagnostic marker panel, improving the efficacy of HCC diagnosis.It is critical to have a thorough grasp of the function of VEGFA in HCC.By elucidating the role of VEGFA and related pathways in HCC, new therapeutic approaches may be developed to inhibit tumor growth and metastasis, which may ultimately improve patient prognosis.
The F13B gene encodes the B subunit of coagulation factor XIII, which is essential for blood clot stabilization (11).Factor XIII is a transglutaminase that cross-links fibrin, thereby strengthening and stabilizing the blood clot (12).The B subunit acts as a carrier for the active A subunit, ensuring its stability and proper function.Mutations in the F13B gene can lead to Factor XIII deficiency, an uncommon bleeding disorder marked by impaired clot formation, delayed wound healing, and spontaneous bleeding (13).Recent research has highlighted the potential involvement of F13B in various cancers.Yang et al. identified F13B as one of the prominently dysregulated proteins in the urine of clear cell renal cell carcinoma (ccRCC) patients, suggesting its potential significance in ccRCC diagnosis and prognosis, although it lacked statistical significance in the final analysis (14).Similarly, Pietsch et al. suggested a potential function for F13B acting as an anti-cancer gene in medulloblastoma pathogenesis, noting that it is located on chromosome 1q31-q32.1,a region exhibiting allelic loss in medulloblastomas (15).Additionally, Iino et al. indicated a plausible association between F13B and colorectal cancer, particularly in the context of DNA microsatellite instability (MSI) in colorectal polyps (16).These studies collectively underscore the potential involvement of F13B in the pathogenesis of various cancers, urging further investigation into its precise mechanisms and clinical implications.Given these findings, it is crucial to explore the function of F13B in HCC to comprehend its possible impact on tumor behavior and patient outcomes.Through the integration of experiments with HCC cell lines and HUVECs, we aimed to explore the gene's intrinsic effects on tumor cells and its impact on angiogenesis, a critical component of tumor progression.This dual approach is designed to gain a deeper understanding of the complex interplay between cancer cells and endothelial cells, which is essential for comprehending metastasis and developing potential therapeutic strategies.
Despite advances in understanding the pathogenesis of HCC, the precise molecular mechanisms remain elusive, highlighting the need for further research.VEGFA is essential for angiogenesis and has a substantial effect on the growth of HCC.Prior research has demonstrated the presence of multiple gene alterations and regulatory pathways in HCC, highlighting the potential for therapeutic interventions targeting these mechanisms (17).The F13B gene has shown potential in different cancers, but its role in HCC has not been fully elucidated.In this investigation, F13B was recognized as a prognostic gene for HCC through integration and bioinformatics analysis.Cell experiments were employed to explore the function of F13B in HCC and its mechanism, focusing on its regulation of VEGFA and HIF-1α/VEGF pathways.We examined the expression of F13B in various HCC cell lines, including Huh-7, SMMC-7721, Hep3B, and MHCC97, each selected for distinct reasons that enhance our research.Our choice of Huh-7 was driven by its high transfection efficiency and stable tumor-forming capabilities, making it ideal for molecular mechanism studies (18).The SMMC-7721 cell line, known for its high metastatic potential, represents a valuable model for understanding HCC's aggressive nature (19).Hep3B was included due to its widespread use in HCC research and unique genetic attributes (20).Lastly, the MHCC97 cell line was chosen for its ability to mimic the invasive and metastatic behavior of HCC (21).By studying the expression and functional effects of F13B in HCC and endothelial cells, we seek to gain insight into its potential as a therapeutic target and prognostic marker, ultimately helping to improve patient outcomes and deepen our understanding of HCC pathogenesis.

Datasets acquisition
A comprehensive search was performed in the Gene Expression Omnibus (GEO) database using the keyword "hepatocellular carcinoma".From this search, we selected the GSE120123 dataset, which included profiles of gene expression of 6 HCC samples and 10 normal control samples.To further validate and comprehensively analyze our findings, we also accessed The Cancer Genome Atlas (TCGA)-Liver hepatocellular carcinoma (LIHC) database (https://cancergenome.nih.gov/).From this database, 371 primary tumor tissues and 50 normal liver tissues were downloaded.These samples served as an independent validation cohort for subsequent gene identification and analysis.The data we utilized have been processed through the HTSeq protocol, ensuring standardized gene expression measurements, which guarantee that our analysis is based on a consistent and reliable dataset.

Differential expression and enrichment analysis
To select differentially expressed genes (DEGs) from the GSE120123 and TCGA-LIHC samples, we utilized the "limma" package (version 3.42.2).The DEG selection criteria were established as |log2 (fold change) FC| < 1 for downregulated DEGs and |log2 (fold change) FC| > 1 for upregulated DEGs, with a significance threshold of P < 0.01.Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses were performed on the GSE120123-DEGs for enrichment using FunRich (http://www.funrich.org/).Three categories were considered in the GO analysis: Cellular Component (CC), Biological Process (BP), and Molecular Function (MF), with findings deemed noteworthy at P < 0.05.

Identification of key overlapping genes among GSE120123-DEGs, TCGA-LIHC-DEGs, and LIHC prognosis-related genes
Subsequently, we utilized the Assistant for Clinical Bioinformation (https://www.aclbi.com/static/index.html#/)database to perform a comprehensive survival analysis on all genes in the TCGA-LIHC dataset, including univariate Cox regression analysis to pinpoint genes significantly linked to overall survival.This is because overall survival is an established and definitive measure in oncology research, providing a clear and direct assessment of the impact of gene expression on patient outcomes.We utilized the "Calculate and draw custom Venn diagrams" tool (https://bioinformatics.psb.ugent.be/webtools/Venn/)for the intersection analysis of prognostic genes and DEGs.This analysis included GSE120123-DEGs, TCGA-DEGs, and prognostic genes identified in the LIHC dataset, allowing us to pinpoint key overlapping genes for further investigation.

Overall survival (OS) analysis and construction of prognostic risk model
Using the Kaplan-Meier (KM) plotter website (https://kmplot.com/),we performed OS analysis of overlapping genes and identified genes with significant results (P < 0.05).Next, prognostic genes were subjected to the Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression analysis using the "glmnet" R package (version 4.1.3).Ten-fold cross-validation analysis was utilized to choose the characteristic gene coefficients based on the minimum value of lambda.Patients were categorized into low-risk and high-risk groups based on the expression of distinctive genes in TCGA-LIHC data sets.The survival status of patients in different risk groups and the clustering of genes were analyzed.To assess survival differences between the two groups, a log-rank test was employed for KM survival analysis.In addition, time-dependent receiver operating characteristic (ROC) analysis was used to evaluate the predictive ability of the risk model, and the area under the curve (AUC) value indicated the discriminative ability of the risk model.

Expression and survival analysis of CCDC170 and F13B
In this study, SangerBox (http://sangerbox.com/) was utilized to examine the levels of two genes (CCDC170 and F13B) between GSE120123 tumors and normal tissues.The impact of differential expression of CCDC170 and F13B on progression-free survival (PFS), diseasespecific survival (DSS), and disease-free interval (DFI) of LIHC patients was then analyzed through the Gene Set Cancer Analysis (GSCA; https://guolab.wchscu.cn/GSCA/#/)database.
Results with P < 0.05 were considered statistically significant.

Construction and validation of predictive nomograms
Uni/multivariate Cox regression analyses between F13B and clinicopathological parameters (age, gender, pT-stage, pTNM-stage) were constructed using the "forestplot" package (version 2.0.1).Following that, nomograms were built using the "rms" package (version 3.6.1).The nomogram provides graphical results that enable the calculation of a patient's prognosis risk.
The clinical parameters included in our nomogram were chosen based on their established prognostic relevance and availability across the patient cohort, ensuring broad applicability and clinical utility.The predictive accuracy of the nomograms was examined using calibration curves, which compare the predicted survival probabilities with the observed outcomes.A closer alignment of the predictive results with the calibration curve indicates better forecasted outcomes of the model.

Gene set enrichment analysis (GSEA) and clinical characteristics analysis
To identify enriched pathways associated with F13B, we performed GSEA to identify biological pathways associated with F13B expression in HCC.UALCAN (http://ualcan.path.uab.edu/analysis.html), an extensive online resource for quick and easy access to TCGA cancer data, was utilized for gene expression analysis and validation.Using the UALCAN dataset, we confirmed the relative expression levels of F13B across different clinical characteristics, including patient age, nodal metastasis status, tumor grade, histological subtypes, and TP53 mutation status.By detecting statistically significant differences in F13B expression across different clinical categories, its potential relevance in various pathological conditions was analyzed.

Cell lines and culture conditions
The Chinese Academy of Sciences Cell Bank (Shanghai, China) provided the HCC cell lines Hep3B, MHCC97, and Huh-7.The SMMC-7721 cell line was obtained from Cobioer Biosciences (Nanjing, China).These cells were cultured in RPMI 1640 medium supplemented with 1% penicillin-streptomycin and 10% fetal bovine serum (FBS) in a humidified atmosphere containing 5% CO2 at 37°C.For hypoxic conditions, HCC cells were transferred to a hypoxia incubator (YCP-80/S, Superhorizon, China) set at 37°C with 1% O2 and 5% CO2 once they reached 70% to 80% confluence.Human umbilical vein endothelial cells (HUVECs) were bought from Cobioer Biosciences (Nanjing, China) and grown in endothelial cell medium enhanced with 1% endothelial cell growth supplement, 1% penicillin-streptomycin, and 5% FBS.To study the function of vascular endothelial growth factor (VEGF) in tumor angiogenesis, VEGF was applied to HUVECs in different concentrations (0, 5, 10, 15, 20 ng/ml) for 24 hours.

Quantitative real-time PCR (qRT-PCR)
TRIzol reagent (Invitrogen) was used to extract total RNA from the cells, following the manufacturer's instructions.A PrimeScript RT kit (TaKaRa) was employed to create complementary DNA (cDNA) from the extracted RNA.SYBR Green PCR Master Mix (TaKaRa) was utilized to conduct quantitative real-time PCR (qRT-PCR) using an ABI 7500 Real-Time PCR System.As an internal control, gene expression levels were normalized to GAPDH.Target gene relative expression was assessed by the 2-ΔΔCT method.The primer sequences used for qRT-PCR are listed in Table 1.

Western blotting (WB)
Identical quantities of protein (30 μg) were separated on a 10% SDS-PAGE gel and transferred onto nitrocellulose membranes.In the knockdown experiment, to ensure that even low levels of F13B could be accurately detected and quantified, we increased the sample loading amount.
After blocking for an hour at room temperature with 5% nonfat milk in TBST buffer, the membranes were incubated overnight at 4°C with primary antibodies against the target proteins.

Transwell assay
For the migration assay, serum-free media containing 2 × 104 HCC cells per well were used in the top chamber of a 24-well Transwell plate, while medium with 10% FBS as a chemoattractant was placed in the lower chamber.The top chamber was coated with Matrigel (BD Biosciences) for the invasion experiment, and 5 × 104 cells were seeded per well in serumfree media.After a 24-hour incubation, non-invaded or non-migrated cells in the top chamber were removed with a cotton swab.Invaded or migrated cells on the membrane's bottom were fixed with 4% paraformaldehyde and stained with DAPI.The stained cells were quantified using a microscope (Olympus, Tokyo, Japan).

Flow cytometry
For flow cytometry analysis of the cell cycle, HCC cells were detached using trypsin-EDTA (Life Technologies Inc., Beijing, China) and washed with phosphate-buffered saline (PBS).
The cells were then fixed overnight at 4°C in 70% ethanol.After washing with PBS, cells were stained with propidium iodide (PI) in the presence of RNase A to quantify DNA content.A flow cytometer (Jiyuan, Guangzhou, China) was used to carry out the flow cytometry.Data were analyzed using FlowJo software (FlowJo, Hangzhou, China) to determine the cell distribution in the G1, S, and G2 phases.

Enzyme-linked immunosorbent assay (ELISA)
The expression of proteins was measured using an ELISA.First, a capture antibody was applied to 96-well plates, which were left to sit at 4°C overnight.The plates were blocked with 5% non-fat milk in PBS for 2 hours.Samples were then incubated in the wells at 37°C for 1 hour.
The detection antibody was added and incubated for 1 hour at 37°C after washing with PBST.
Following this, a secondary HRP-conjugated antibody was incubated at 37°C for another hour.
The substrate solution was added, and the reaction was stopped after 15 minutes with a stop solution.Finally, the absorbance at 450 nm was measured using a microplate reader (Thermo Scientific, Shanghai, China).

Cytotoxicity assay
Lactate dehydrogenase (LDH) activity was assessed to evaluate cellular cytotoxicity.HUVECs were seeded in 96-well culture plates and treated with various concentrations of VEGF (0, 5, 10, 15, or 20 ng/ml) for 24 hours.Cellular proteins were extracted and analyzed using a commercially available LDH activity assay kit from Jiangsu KeyGen BioTECH Corp., Ltd.

Ethics approval and consent to participate
Not applicable.

Statistical analysis
Statistical analysis was performed using the R programming language.All experiments were conducted in triplicate, and results are expressed as mean ± standard deviation (SD).One-way ANOVA was used to determine significance between multiple groups, with Tukey's post-hoc test applied for pairwise comparisons.For analyses involving multiple variables, two-way ANOVA was used, with Bonferroni correction applied for multiple comparisons in post-hoc tests.A P-value of < 0.05 was considered statistically significant.

Classification and functional enrichment analysis of GSE120123-DEGs
From the GSE120123 dataset, 1372 DEGs were identified, including 1107 upregulated and 265 downregulated DEGs (Figure 1A).The cluster distribution of these DEGs in the dataset is shown in Figure 1B.The GO analysis revealed that these DEGs were enriched in terms like receptor ligand activity, cellular response to cytokine, and tertiary granule lumen (Figure 1C).
KEGG pathway analysis showed significant enrichment in cytokine-cytokine receptor interaction, IL-17 signaling pathway, neuroactive ligand-receptor interaction, hematopoietic cell lineage, and viral protein interaction with cytokine and cytokine receptor pathways (Figure 1D).

Screening and survival prognostic analysis of 220 overlapping genes
In the TCGA-LIHC database, comprehensive OS analysis of all genes identified 16,491 prognosis-related genes, with the top 20 highlighted in Figure 2A.Differential expression analysis of TCGA-LIHC samples identified 2,558 upregulated and 934 downregulated DEGs (Figure 2B).Intersection analysis of prognostic genes, GSE120123-DEGs, and TCGA-LIHC-DEGs identified 220 overlapping genes for further investigation (Figure 2C).OS analysis on these genes revealed 57 with significant P-values (Supplementary Figure 1).

LASSO-Cox regression analysis and prognostic model validation
LASSO-Cox regression analysis was performed on the 57 genes with significant P-values.
Through ten-fold cross-validation, we identified eight key genes with λmin = 0.0429 (Supplementary Figures 3A and 3B).The risk score formula for these genes was established as groups based on gene expression.The risk score's association with survival time/status for these genes is shown in Supplementary Figure 2C.KM analysis indicated that OS probability was lower in high-risk patients compared to low-risk patients (Supplementary Figure 2D).The predictive performance of the prognostic model was evaluated using ROC curve analysis.AUC values for predicting 1-, 3-, and 5-year outcomes were 0.839, 0.773, and 0.702, respectively, indicating significant prognostic value (Supplementary Figure 2E).

Expression analysis and prognosis assessment of key genes
Among the eight prognostic genes, CCDC170 and F13B had not been previously reported in HCC.We selected these two genes for further analysis.According to Supplementary Figure 3A, CCDC170 expression was significantly lower in tumor samples than in normal samples (***P < 0.001).Survival analysis for CCDC170 showed that lower expression was linked to worse PFS (P = 0.042) and DSS (P = 0.037), but there was no significant correlation with DFI (P = 0.39) (Supplementary Figures 3B-3D).Similarly, F13B expression was lower in tumor samples compared to normal samples (Supplementary Figure 3E).Lower F13B expression was associated with worse PFS (P = 0.0038), DSS (P = 0.038), and DFI (P = 0.027) (Supplementary Figures 3F-3H).These findings suggest that F13B is a significant prognostic factor in LIHC and warrants further investigation.

Prognostic validation of F13B and nomogram construction for survival prediction
Univariate and multivariate Cox regression analyses validated the prognostic significance of F13B.In the univariate analysis (Figure 3A), F13B expression was strongly associated with OS (P = 0.00246, HR = 0.87245, 95% CI: 0.79781-0.953).Multivariate analysis confirmed that F13B was an independent prognostic factor for OS (P = 0.03429, HR = 0.90504, 95% CI: 0.82517-0.99264)(Figure 3B).A nomogram was constructed to predict 1-year, 3-year, and 5-year survival probabilities based on F13B expression (Figure 3C).The C-index for the nomogram was 0.565 (95% CI: 0.504-1), and the P-value was 0.038, indicating moderate predictive accuracy.Calibration curves showed a good alignment between predicted and observed survival probabilities at 1, 3, and 5 years (Figure 3D).These results underscore the prognostic value of F13B in HCC, supporting its role as a reliable biomarker for predicting patient outcomes.

Correlation of F13B expression with clinical characteristics in LIHC
GSEA-enriched pathways for F13B included genes controlling nephrogenesis, the PPAR signaling pathway, and EGFR tyrosine kinase inhibitor resistance (Supplementary Figure 4A).
F13B mRNA expression was analyzed across different clinical subgroups based on patient age, TP53 mutation status, nodal metastasis status, tumor grade, and histological subtypes.F13B expression increased with patient age (Supplementary Figure 4B), decreased with higher tumor grades (Supplementary Figure 4C), and was higher in the N0 nodal metastasis group (Supplementary Figure 4D).TP53-NonMutant patients exhibited higher F13B expression than TP53-Mutant patients (Supplementary Figure 4E).Among different histological subtypes, F13B expression was highest in hepatocellular carcinoma (Supplementary Figure 4F).

F13B regulates the proliferation, invasion, migration, and cell cycle of HCC cells
In cell line experiments, F13B expression levels were found to be significantly reduced in HCC cell lines, particularly in SMMC-7721 and Huh-7 (Figures 4A and 4B).The knockdown and overexpression efficiency of F13B in SMMC-7721 and Huh-7 cells was confirmed, with si-F13B#1 showing the most significant knockdown efficiency (Figures 4C-4J).CCK-8 assays revealed that F13B overexpression inhibited SMMC-7721 cell proliferation, while knockdown enhanced proliferative capacity (Figures 4K-4N).Transwell assays showed that F13B overexpression impeded SMMC-7721 cell migration and invasion, while knockdown promoted these processes (Figures 5A-5D).Flow cytometry analysis indicated that F13B overexpression reduced the proportion of cells in the S phase and increased the proportion in the G1 phase, while F13B knockdown caused the opposite effect (Figures 5E-5H).
To investigate the molecular mechanisms by which F13B regulates the cell cycle, we conducted qRT-PCR and WB assays to assess the expression levels of key cell cycle-related proteins, including CDC2, Cyclin B1, CDC25, p21, and Cyclin D1, in SMMC-7721 and Huh-7 cells.
Overexpression of F13B significantly reduced the expression levels of Cyclin B1, CDC2, CDC25, and Cyclin D1 while increasing p21 levels (Figures 6A-6F).Conversely, F13B knockdown had the opposite effects, increasing Cyclin B1, CDC2, CDC25, and Cyclin D1 while reducing p21 expression (Figures 6G-6L).These findings suggest that F13B plays a crucial role in regulating cell cycle progression, likely through the modulation of these key proteins.

Knockdown of F13B promoted the proliferation and migration of HUVEC
HUVEC is commonly used as a model for studying angiogenesis (25).In this study, we analyzed the effects of F13B knockdown or overexpression on the expression of the microvascular marker CD31 in HUVEC via Western blot (WB) experiments.Our results demonstrated that F13B knockdown led to an upregulation of CD31 protein levels, while overexpression of F13B caused a decrease in CD31 expression (Figures 7A-7C).Functional assays further revealed that F13B overexpression inhibited HUVEC proliferation and migration, while knockdown of F13B had the opposite effect (Figures 7D-7G).These findings suggest that F13B plays a crucial role in regulating HUVEC angiogenesis.

Overexpression of F13B inhibits VEGF-induced proliferation and invasion of HUVEC
The LDH activity assay is commonly used to evaluate cell membrane integrity and cytotoxicity (26).When cell membranes are compromised, LDH is released into the culture medium, and an increase in LDH activity is indicative of cytotoxicity.Using an LDH detection kit, we assessed the impact of different concentrations of VEGF (0, 5, 10, 15, 20 ng/ml) on HUVEC cytotoxicity after 24 hours.A significant decrease in cytotoxicity was observed only at 20 ng/ml VEGF (Figure 8A).Additionally, CCK-8 and transwell assays showed that 20 ng/ml VEGF significantly increased the viability and invasive capacity of HUVEC cells, but this increase was attenuated by F13B overexpression (Figures 8B and 8C).

F13B inhibits HUVEC proliferation by downregulating the expression of VEGFA
VEGFA is a major angiogenic factor critical in tumor angiogenesis (27).We examined the regulation of VEGFA by F13B in HUVEC cells.ELISA results indicated that F13B overexpression suppressed VEGFA secretion in the HUVEC supernatant (Figure 9A), while F13B knockdown promoted it (Figure 9B).Subsequent qRT-PCR and WB experiments confirmed that F13B negatively regulates VEGFA expression at both the mRNA and protein levels (Figures 9C-9H).F13B overexpression also inhibited HUVEC proliferation, while overexpression of VEGFA rescued this inhibition (Figure 9I).Conversely, knockdown of F13B stimulated HUVEC growth, but this effect was reversed by VEGFA knockdown (Figure 9J).
Overexpression of F13B reduced the levels of these proteins.Additionally, WB analysis showed that VEGF treatment raised p-mTOR and p-AKT protein levels in HUVEC, but F13B overexpression alleviated this increase (Figures 10E-10G).These findings indicate that F13B suppresses angiogenesis by inhibiting the AKT/mTOR signaling pathway and downregulating VEGFR2, MMP2, and MMP9.

F13B affects HCC through the HIF-1α/VEGF pathway
We also examined the effect of F13B on VEGF expression in SMMC-7721 and Huh-7 cells.
qRT-PCR and WB results showed that hypoxic conditions significantly increased VEGF expression, which was reduced by F13B overexpression (Figures 11A-11E).Under hypoxic conditions, cell viability increased, but this effect was mitigated by F13B overexpression (Figures 11F and 11G).HIF1A, a key regulator of the cellular response to low oxygen, is known to upregulate VEGF expression.Our analyses revealed that F13B overexpression also downregulated HIF-1α levels in hypoxic conditions (Figures 11H-11L), suggesting that F13B modulates HCC progression through the HIF-1α/VEGF pathway.

DISCUSSION
HCC is one of the main causes of cancer-related deaths worldwide, characterized by its aggressive nature and poor prognosis (28).Understanding the molecular mechanisms underlying HCC is essential for developing effective treatments and therapeutic strategies.In this study, we screened DEGs from the GSE120123 dataset and identified significant enrichment in terms related to inflammatory response, the IL-17 signaling pathway, and hematopoietic cell lineage.Prior research has highlighted the importance of these pathways in HCC.For example, Yang AY et al. suggested that Scutellaria barbata may treat HCC by inhibiting core genes and blocking the IL-17 signaling pathway, thereby suppressing cancer cell proliferation and migration while inducing apoptosis (29).Similarly, Yu L et al. revealed that NF-kappaB plays a role in regulating Tec, a protein tyrosine kinase, suggesting its potential association with HCC development via the hematopoietic lineage pathway (30).
We conducted univariate Cox analysis on the TCGA-LIHC dataset to identify genes associated with prognosis.The TCGA-LIHC dataset was chosen for its comprehensive clinical data and rigorous quality control, making it a reliable resource for identifying prognostic genes.
Overlapping genes from prognostically significant genes, DEGs of TCGA-LIHC, and DEGs from the GSE120123 dataset were screened for overall survival analysis, resulting in 57 genes with significant P-values.Among the 8 prognostic feature genes, CCDC170 and F13B have not been previously reported in HCC.F13B, in particular, was selected as a key gene after the prognostic risk model, expression, and survival analysis.Pathway enrichment analysis using GSEA revealed that F13B was significantly enriched in pathways related to nephrogenesis, the PPAR signaling pathway, and EGFR tyrosine kinase inhibitor resistance (21,31,32).These In summary, our findings suggest that the IL-17 signaling, EGFR inhibitor resistance, and PPAR pathways are crucial in HCC development.Identifying enriched genes within these pathways could provide promising directions for HCC diagnosis and therapy.
The cell cycle is a tightly regulated process that governs cell growth and division (33).Key regulators include Cyclin B1, CDC2, CDC25, p21, and Cyclin D1.Cyclin B1 complexes with CDC2 to promote the transition from the G2 phase to mitosis, while CDC25 activates CDC2 by removing inhibitory phosphates (34).p21, a CDK inhibitor, blocks Cyclin-CDK complexes, causing cell cycle arrest (35).Cyclin D1 controls the transition from the G1 to the S phase, promoting cell cycle progression (36).Together, these molecules ensure accurate cell division and proliferation.In HCC, dysregulation of the cell cycle is associated with poor prognosis and tumor progression.Zuo SR et al. showed that SMYD2 expression in HCC correlates with aggressive tumor features and poor prognosis (37).Silencing SMYD2 inhibits cell proliferation and cycle progression, highlighting its potential as a prognostic biomarker.Similarly, Zuo J et al. demonstrated that TCP10L has tumor-suppressive properties in HCC, with its downregulation associated with advanced disease stages (38).Overexpression of TCP10L reduces colony formation, suggesting its potential as a diagnostic and therapeutic target.
Additionally, Zou X et al. found that knockdown of lncRNA HUMT inhibits HCC proliferation and metastasis through the miR-455-5p/LRP4 axis, positioning it as a promising therapeutic target (39).Our in vitro experiments showed that F13B expression is low in HCC cell lines and tumor tissues.F13B overexpression inhibits invasion, migration, and proliferation of HCC cells while also affecting the expression of cell cycle proteins, leading to cell cycle arrest.These findings suggest that F13B could serve as an independent biomarker for HCC prognosis and a potential therapeutic target.
HUVECs are commonly used as a model for studying angiogenesis.By integrating experiments with both HCC cell lines and HUVECs, we aimed to explore F13B's intrinsic effects on tumor cells and its impact on angiogenesis, a critical factor in tumor progression.This dual approach helps provide a comprehensive understanding of the interplay between cancer cells and endothelial cells, which is crucial for metastasis and the development of therapeutic strategies.
VEGF is essential for angiogenesis as it promotes endothelial cell migration, proliferation, and tube formation.When exposed to VEGF, HUVECs activate VEGF receptors, triggering signaling cascades that upregulate angiogenesis-related genes, which are vital for wound healing and tissue regeneration (40) (43).CD31 is a marker for endothelial cells used to identify microvascular density, correlating with angiogenesis in tumors (44).Increased CD31 expression indicates enhanced angiogenesis, often observed in aggressive tumors (42).Our study found that F13B inhibits CD31 expression in HUVECs, suggesting it suppresses HCC angiogenesis.When HUVECs were exposed to VEGF, LDH release was reduced, and cell proliferation and invasion were enhanced.However, these effects were mitigated by F13B overexpression.LDH release is a marker of cell damage and cytotoxicity, and reduced LDH release upon VEGF induction indicates lower cell damage and higher viability, which aligns with increased proliferation and invasion.Knockdown of F13B in HCC cells increased VEGFA secretion and expression, while F13B overexpression inhibited HUVEC proliferation-an effect reversed by VEGFA overexpression.These findings suggest that F13B inhibits HCC angiogenesis by downregulating VEGFA expression, making it a potential therapeutic target in HCC.
The AKT/mTOR pathway is a key signaling cascade regulating cell growth, survival, and metabolism (45).Activation starts when growth factors bind to cell surface receptors, leading to AKT phosphorylation.Activated AKT phosphorylates and activates mTOR, a central kinase that controls protein synthesis and cell growth (46).The AKT/mTOR pathway integrates signals from upstream pathways like PI3K and PTEN, playing a vital role in cell homeostasis.
This pathway is also important in cancer development.*P < 0.05 vs. control, #P < 0.05 vs. vector+Hypoxia.

Figure 1 .
Figure 1.Identification and functional enrichment analysis of DEGs in the GSE120123 dataset.(A) Volcano map, the distribution of DEGs in the GSE120123 dataset.Red dots represent 1107 up-regulated DEGs, blue dots represent 265 down-regulated DEGs, and gray dots represent genes that did not meet the screening criteria.(B) Heatmap of expression of DEGs in the GSE120123 dataset.Pink columns represent tumor

Figure 2 .
Figure 2. Prognostic gene analysis in the TCGA-LIHC dataset.(A) Univariate Cox analysis in the TCGA-LIHC dataset, forest plot showing the top 20 genes associated with LIHC prognosis.(B) Heat map of DEGs based on the TCGA-LIHC database.Blue represents the control group, and yellow represents the tumor group.

Figure 3 .
Figure 3. Determination of independent prognostic parameters and construction of genebased prognostic models.(A and B) Forest plots of univariate and multivariate Cox proportional hazards regression analyses.HR and corresponding 95% CIs for the association between the independent variable F13B and OS, taking into account other clinical variables including age, sex, pT stage, pTNMstage, and grade.Each row represents a clinical variable, and horizontal lines represent null values (HR=1).Points to the left of the line indicate higher risk (HR>1), while points to the right of the line indicate lower risk (HR<1).(C) Nomogram of F13B predicting 1-, 3-, and 5-year OS in HCC patients.The line segment corresponding to the variable represents the range of possible values of the variable, and the length of the line segment reflects the contribution of the variable to the final event.(D)Calibration plots of nomograms for consistency tests between 1-, 3-, and 5-year OS

Figure 4 .
Figure 4. F13B knockdown enhances the proliferation of HCC cells.(A and B) Relative F13B mRNA and protein levels in normal liver cell line (THLE-2) and HCC cell lines (Huh-7, SMMC-7721, Hep3B, and MHCC97) as determined by qRT-PCR and WB

Figure 5 .
Figure 5. Functional effects of F13B on cell invasion, migration, and cell cycle progression.(A-D) Transwell assay results showing the effects of over-F13B/si-F13B#1 on the invasion and migration capacity of SMMC-7721 and Huh-7 cells.The bar graph represents the quantification of invaded and migrated cells.(E-H) Flow cytometry analysis revealed cell cycle distribution of SMMC-7721 and Huh-7 cells after F13B overexpression/knockdown.The bar graph represents the percentage of cells in G1, S, and G2 phases.*P < 0.05.Scale bar: 50 μm.

Figure 6 .
Figure 6.Effects of knockdown or overexpression of F13B on cyclin expression in HCC.(A-F) After overexpression of F13B, changes in cyclin expression in HCC cells were measured by qRT-PCR and WB.Cyclins include Cyclin B1, CDC2, CDC25, p21, and Cyclin D1.(G-L)After knocking down F13B, changes in cyclin expression in HCC cells were measured by qRT-PCR and WB.HCC: hepatocellular carcinoma, qRT-PCR: Quantitative real-time polymerase chain reaction,

Figure 9 .
Figure 9. F13B regulates VEGFA expression and affects proliferation of HUVEC cells.(A) ELISA analysis showing the secretion of VEGFA in the supernatant of HCC cells with F13B overexpression.(B) ELISA analysis demonstrating the secretion of VEGFA in the supernatant of HCC cells with F13B knockdown.(C-E) qRT-PCR and WB to examine the expression level of VEGFA mRNA/protein in HCC cells after F13B overexpression.(F-H) qRT-PCR and WB to examine the expression level of VEGFA mRNA/protein in HCC cells after F13B knockdown.(I and J) CCK-8 detection of HUVEC proliferation activity overexpressing F13B and VEGFA or knocking down F13B and VEGFA.HUVEC: human umbilical vein endothelial cells, ELISA: Enzyme-linked immunosorbent assay, HCC: hepatocellular carcinoma, qRT-PCR: Quantitative real-time polymerase chain reaction, WB: Western blotting, CCK-8: Cell Counting Kit-8.*P<0.05.

Figure 11 .
Figure 11.F13B affects HCC through the HIF-1α/VEGF pathway.(A-E) WB detection of VEGF expression in SMMC-7721 and Huh-7 cells after overexpression of F13B in the presence or absence of hypoxic environment.(F and G) CCK-8 detects the viability of SMMC-7721 and Huh-7 cells after overexpressing F13B with or without hypoxia.(H-L) qRT-PCR and WB detected the expression of HIF-1α in SMMC-7721 and Huh-7 cells after overexpressing F13B with or without hypoxia.HCC: hepatocellular carcinoma, qRT-PCR: Quantitative real-time polymerase chain reaction, WB: Western blot, CCK-8: Cell Counting Kit-8.
. Zhu Z et al. showed that NT5DC2 knockdown reduces VEGF expression in colorectal cancer, hindering tumor growth and metastasis (41).Similarly,