All the gene expression profile files and the clinical information were obtained from two public databases. TCGA (The Cancer Genome Atlas database, https://portal.gdc.cancer.gov/) contains information on transcriptional gene expression in human cancer and healthy tissues. After removing samples with incomplete clinical information, 369 liver tumor tissues and 50 healthy liver tissues were analyzed. In the GEO (Gene Expression Omnibus, www.ncbi.nlm.nih.gov/geo/), incompletely annotated samples were excluded from the dataset of GSE14520 (sequencing platform: GPL3921). The analysis included 210 samples, comprising both tumor samples and their corresponding non-cancer pairs. The clinical information of all the samples is presented in Table 1. Liver tumor tissue microarrays of nine paired samples (HLiv-H030PG03) were purchased from Shanghai Xinchao (Shanghai, China).

TCGA data were converted to TPM (transcripts per million) values for subsequent analyses. For GSE14520, data normalization and log2 transformation were performed. Probe IDs were converted to gene symbols using a platform annotation file. Genes with multiple probes are represented by their maximum expression values. Data pre-processing was performed, followed by differential expression and statistical analyses using R software. A survival curve analysis was performed using the R packages "Survival" and "Survminer" with the best cut-off value for MTDH calculated for both datasets. R packages “ggstatsplot” and “corrplot” were used for data visualization.

Huh7 cells were purchased from the Chinese Academy of Sciences Cell Bank and, MHCC-97H cells were obtained from the Liver Cancer Institute at Fudan University Zhongshan Hospital. All the cells were cultured in DMEM medium [10% fetal bovine serum (FBS)], 100 U/mL penicillin, and 100 U/mL streptomycin in an incubator with 5% CO₂.

Tumor spheres were cultured in serum-free medium (SFM) consisting of 20 ng/mL EGF (PeproTech), DMEM/F12 (HyClone), 20 ng/mL bFGF (PeproTech), and 20 μL/mL B27 (Gibco). Single cells (1 × 10⁴) were seeded in a 6-well ultra-low adsorption plate (Corning) containing 1.5 mL of serum-free culture medium. All the cells were incubated for 14 d in incubators with 5% CO₂ at 37°C. Inverted microscopes were used to count tumor spheres larger than 50 mm.

To prepare proteins from the cells, the cells were lysed in RIPA solution (CWBIO) containing a protease inhibitor cocktail solution. The protein concentration in the solution was determined using a BCA kit (Beyotime). Proteins were separated by using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (Beyotime Millipore) and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). PVDF membranes were closed with 5% skim milk at room temperature (RT) for 1 h, followed by overnight incubation with primary antibody at 4°C. The membranes were washed thrice with TBST containing 0.1% Tween-20, followed by incubation with secondary antibodies (EarthOx) (1:5000) for 1 h. After washing thrice with TBST, for 5 min, the blots were visualized with chemiluminescence reagents (Beyotime). The antibodies used were anti-MTDH (CST, 14065T), anti-CD133 (YT5192, ImmunoWay), anti-NANOG (CST, 4903T), and anti-GAPDH (YM3215, ImmunoWay).

TRIzol (Takara) was used to extract RNA from the cells, and the cDNA (complementary DNA) composition was determined using the PrimeScript^TM RT Reagent Kit (TaKaRa). Real-time quantitative reverse transcription PCR (qRT-PCR) was performed using SYBR Premix Ex Taq (TaKaRa Bio). A CFX96 Real-Time PCR Detection System (Bio-Rad) was used for qRT-PCR analysis. PCR was performed using the primers purchased from Sangon Biotechnology (Shanghai, China). The ^2-ΔΔCt method was used to calculate the data. Table 2 shows primer sequences.

To establish a cell line with stable overexpression and suppression of MTDH, the cell were infected with a lentivirus. MTDH-overexpressing lentiviral vectors, MTDH-RNAi lentiviral vectors, and blank control lentiviral vectors were acquired from GeneChem (Shanghai, China). On the previous day, 3 × 10⁴ cells were grown in each well of a six-well plate. When the cells reached 30% confluence, they were transfected according to the manufacturer’s protocols at Multiplicity of Infection of 5. After 96 h, fluorescence microscopy and western blotting were used to verify the transfection efficiency. Puromycin screening was then performed for 2 wk (GeneChem).

The cells were seeded in a 24-well plate. Paraformaldehyde solution (4%) was used to fix the cells for 20 min after 24 h of incubation. For cell membrane permeabilization, the cells were treated with 0.3% TritonX-100 (Sigmautes) for 15 min. The cells were incubated overnight with anti-NANOG antibody at 4°C after 120 min incubation in 5% BSA in TBST. Next, the cells were washed thrice with PBS and incubated with DyLight 649 AffiniPure Goat Anti-Rat IgG (1:200, Abbkine) for 1 h in a dark humidified box. The final counterstaining was done using DAPI (4',6-diamino-2-phenylindole) for 10 min. Images were captured using a fluoresce microscope (Nikon).

MHCC-97H and Huh7 cells were cultured in six-well plates. Subsequent experiments were performed when the cell density in the wells reached 90%. After scratching the cells with a 200-mL pipette, 2% FBS cell culture medium was used for continued culture. All the cells were incubated with 5% carbon dioxide in an incubator at 37°C. A microscope was used to photograph the wounded areas.

The migration and invasion abilities of the cells were assessed using 24-well transwell chambers with or without Matrigel (BD Biosciences). The supplements were kept in 5% CO₂ at 37°C for 12 h. Next, 2 × 10⁵ cells were re-suspended in 200 μL of SFM and placed in transwell cups with an 8-micron pore membrane (BD). Simultaneously, cell culture medium containing 10% FBS was added to the lower layer to make a total volume of 600 μL. The chambers were maintained in 5% CO₂ for 24 h at 37°C. The cells that failed to pass through the pores and remained in the upper chamber were carefully wiped off with a cotton swab. Migrating cells were fixed in 4% paraformaldehyde solution for 20 min. Next, the cells remaining in the chambers were stained in 200 μL 0.1% crystal violet for 30 min, followed by counting with a microscope in five random areas.

The cells were washed with PBS after trypsin digestion. Next, they were suspended in 80 μL of PBS, followed by the addition of 2 μL PE-CD133 antibody (Miltenyi Biotec) and 20 μL of FcR Blocking Reagent (Miltenyi Biotec). Under light-proof conditions, the mixture was incubated at 4°C for 10 min. Next, all the cells were washed thrice in PBS, followed by resuspension with 500 μL of PBS. The mixture was analyzed using a FACS Calibur Flow cytometer (BD Biosciences).

BALB/c-nu mice were obtained from the Animal Experiment Center of Chongqing Medical University. The experiments were conducted in the animal facilities of the Animal Experiment Center of Chongqing Medical University (12-h light/dark cycle, 27 ± 2°C, 50% ± 10% humidity). Nude mice were randomly divided into two groups of three mice each. MHCC-MTDH-LV cells (2 × 10⁶ cells) were re-suspended in a complete culture medium. Next, 100 μL of cell suspension were subcutaneously injected into nude mouse (female, 4–6 wk old). The tumors were detached after 4 wk. Next, the tumor volume was calculated using the following formula: Length × width²/2. Tumor tissues were subjected to immunohistochemical staining (IHC) analysis. This study was approved by the Ethics Committee of the Second Hospital of Chongqing Medical University.

All the slides were dewaxed and gradually hydrated. Antigen extraction was performed using citrate buffer under high pressure and temperature conditions. All the slides were rinsed with PBS and incubated with goat serum for 10 min at RT. Next, the slides were exposed to anti-MTDH and anti-CD133 at 4°C overnight. Subsequently, horseradish peroxidase-labeled streptavidin was added, followed by the addition of DAB reagent. Tap water was dehydrated using an ethanol gradient. The slides were then washed twice with xylene and wrapped in neutral balsam. Protein expression levels were assessed using the ImageJ software. The average optical density was calculated by measuring the integrated optical density and the area of each image, which reflected the concentration per unit area of the target protein.

The infiltration abundances of 28 immune cell species in TCGA samples were quantified using ssGSEA. The gene set data for the immune cells were downloaded from the website (https://www.cell.com/cms/10.1016/j.celrep.2016.12.019/). Box plots were constructed to compare immune cell differences between the groups. The R package used was “GSVA”[18], “ggplot2.”

TISIDB is an online analysis site for the analysis of tumor-immune interactions (http://cis.hku.hk/TISIDB/)[19]. To further investigate the immunological impact of MTDH in cancer, we analyzed and evaluated the TISIDB database through the "Immunomodulators" and “chemokine modules”.

All the experiments were conducted at least thrice. Data were analyzed using R 4.1.2 and GraphPad Prism software. Student's t-test was used to compare two groups of continuous variables. The Kruskal-Wallis and Wilcox tests were used for non-parametric tests. A P value < 0.05 indicated statistical significance.

To identify aberrant MTDH expression in HCC, we downloaded microarray gene profiling data from GEO (GSE14520/GPL3921) and TCGA. In 50 normal liver tissues and 369 Liver cancer tissues from TCGA, MTDH mRNA expression was upregulated in liver tumor tissues. Similarly, MTDH expression was increased in the tumor in GSE14520 (n = 420; Figure 1A and B). MTDH expression was examined in human HCC (n = 9) and paracancerous tissues (n = 9) using IHC. According to these results, HCC tissues overexpressed MTDH, compared with the para-cancerous tissues (Figure 1C). Furthermore, we constructed the Kaplan-Meier curve of HCC using “Survival” and “Survminer” R packages. Shorter survival times were associated with higher MTDH (Figure 1D and E). A higher expression of MTDH was observed in liver cancer tissues than in normal liver tissues or para-cancerous tissues, which had impact on the prognosis of patients with liver cancer.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Metadherin overexpression has been linked to a worse prognosis in hepatocellular carcinoma. A and B: Metadherin (MTDH) mRNA expression in normal liver tissues compared with liver cancer tissues; C: Images of para-carcinoma (n = 9) and cancer tissues (n = 9) stained with MTDH by Immunohistochemical staining (scale bar = 200 μm); D and E: Patient's overall survival curves according to MTDH expression. ^aP < 0.0001, ^bP < 0.01. Normal: Normal liver tissues; Cancer: Liver cancer tissues; Para-carcinoma: Para-cancerous tissue; MTDH: Metadherin; IHC: Immunohistochemical staining.

Two cell lines, Huh7 and MHCC-97H, were transfected with the related lentiviruses. Specifically, we used a blank control (LV-NC) vs overexpression (LV-OE) and a blank control (LV-NC) vs knockdown (LV-RNAi), to obtain stable overexpression and suppression of MTDH. Western blotting was performed to verify MTDH protein expression after transfection (Figure 2A). The migration and invasion assays showed that MTDH overexpression significantly accelerated the migration and invasion of MHCC-97H and Huh7 cells (Figure 2B). Conversely, MTDH knockdown cells exhibited lower potential for migration and invasion (Figure 2C). The wound healing assay confirmed that MTDH overexpression significantly promoted wound healing, whereas MTDH knockdown suppressed scratch wound healing in HCC cells (Figure 2D). These results indicate that high MTDH expression correlates positively with the migration and invasive abilities of HCC cells.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Metadherin promotes proliferation, migration, and invasion of hepatocellular carcinoma cells. A: In Huh7 and MHCC-97H cells, Western blotting revealed the efficacy of Metadherin overexpression and knockdown; B and C: Characteristic images of trans-well invasion assays 24 h after culture; D: Images depicting scratch width at 0 h and 48 h post-scratch in cells captured using inverted microscopy. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001, ^dP < 0.0001. OE: Metadherin overexpression group; NC: Overexpression control group; RNAi: Metadherin knockdown group; RNAi-NC: MTDH knockdown control group.

To determine whether MTDH is related to the stem cell phenotype, we examined MTDH and HCC stemness using TCGA database. TCGA data showed a significant positive correlation between MTDH expression in HCC tissues and CD133, NANOG, and Oct4 expression (n = 369; Figure 3A). Our previous research confirmed that liver CSCs (LCSCs) can be enriched using a serum-free stem cell medium to promote tumor sphere formation. Next, we performed SFM on MHCC-97H and Huh7 cells, followed by western blotting and qRT-PCR to determine MTDH expression in stem cell spheres and attached cells. Compared with liver cancer-adherent cells, LCSCs expressed increased mRNA levels of MTDH and the stem cell markers (CD133, Oct4, and Nanog; Figure 3B). The results showed that the protein expression levels of MTDH, CD133, and NANOG were higher than those in liver cancer adherent cells (Figure 3C). These results showed that MTDH was highly expressed in LCSCs.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Metadherin correlates with the stemness properties in hepatocellular carcinoma. A: Correlation between Metadherin (MTDH) and CD133, Nanog, Oct4 in TCGA; B: Through quantitative reverse transcription PCR, the expression levels of MTDH, CD133, Nanog, Oct4 were measured in attached cells and tumor spheres in 97H and Huh7 cell lines. In tumor spheres, all the four genes were expressed at increased levels; C: MTDH, CD133, and Nanog were higher in attached cells than in tumor spheres. Gene GAPDH served as the internal reference. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001, ^dP < 0.0001.CSC: Cancer stem cells; CC: Cancer cells.

TTo confirm that MTDH maintained stem-like phenotypes in HCC cells, stem cell markers were detected in MHCC-97H and Huh7 cell lines overexpressing MTDH (CD133, Oct4, Nanog). Through PCR experiments, we demonstrated that the markers of CSCs markers were higher in MTDH-overexpressing cells than in the NC groups (Figure 4A). MTDH-overexpressing Huh7 and MHCC-97H cells also expressed high levels of CD133 and Nanog protein (Figure 4B). A The sphere culture assay showed the formation of more spheres in MTDH-overexpressing Huh7 and MHCC-97H cells than in control cells (Figure 4C). Flow cytometry revealed that MTDH overexpression effectively increased the number of CD133⁺ HCC cells (Figure 4D). In addition, MTDH overexpression effectively enhanced CSCs phenotypes in HCC cells.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Metadherin overexpression promotes stem cell phenotypes and self- renewal in hepatocellular carcinoma cell lines. A and B: Through quantitative reverse transcription PCR polymerase chain reaction and Western blot, we determined Metadherin expression and stemness markers; C: The typical pictures of sphere formation assays from 97H-overexpression and Huh7-overexpression cells; D: Flow cytometric analysis of CD133⁺ cells in 97H-overexpressing and Huh7-overexpressing cells. ^aP < 0.05, ^bP < 0.01. OE: Metadherin overexpression group; NC: Overexpression control group; MTDH: Metadherin.

The PCR results demonstrated that the mRNA expression of stem cell markers in MTDH-RNAi cells was lower than that in the NC group (Figure 5A). In contrast to the NC group, MTDH-RNAi cells showed lower expression of CD133 and Nanog (Figure 5B). The downregulation of MTDH reduced the numbers of both MHCC-97H and Huh7 cell spheres, compared with those in the control LV-NC group, as indicated by the results of the sphere culture assay (Figure 5C). The results of the immunofluorescence showed that Nanog fluorescence intensity was greater in the NC group than in the MTDH-RNAi group (Figure 5D). Taken together, these results confirm that the inhibition of MTDH expression attenuates the acquisition of the HCC stem cell phenotype.

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Metadherin downregulation inhibits hepatocellular carcinoma stem cell phenotypes. A: mRNA expression of Metadherin (MTDH), CD133, Nanog, and Oct4 in 97H and Huh7-RNAi; B: In comparison with that of MTDH-LV-RNAi, the protein expression of CD133 and Nanog was elevated in MHCC-97H-NC and Huh7-NC cells; C: Images depicting sphere formation by 97H-RNAi and Huh7-RNAi cells; D: Immunofluorescence images of Nanog (red) in 97H-LV and Huh7-LV samples. DAPI (blue) was used to stain the nuclei. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. RNAi: Metadherin knockdown group; RNAi-NC: Metadherin knockdown control group.

Male BALB/c nude mice were injected with LV-NC or LV-MTDH-RNAi MHCC-97H cells to evaluate the effect of MTDH on tumor growth. The MTDH-RNAi group exhibited a smaller tumor volume than the 97H-NC group, confirming that MTDH promotes tumor growth (Figure 6A and B). Images of hematoxylin and eosin-stained tumor tissues of nude mice are shown in Figure 6C. To further clarify the role of MTDH in the tumor stem cell phenotype, CD133 and MTDH proteins were detected by IHC in the tumor tissues of nude mice. We found that CD133 expression decreased with reduced MTDH expression (Figure 6D). Therefore, these results further suggest that MTDH overexpression promotes liver tumor growth and the CSCs phenotype.

Open in New Tab   Full Size Figure   Download Figure

Figure 6 Metadherin stimulates tumorigenesis in vivo. A: Tumors derived from nude mice injected with 97H-RNAi (n = 3), 97H-NC cell (n = 3); B: Tumor volume showed that the inhibition of Metadherin (MTDH) significantly inhibited tumor growth; C and D: Representative immunohistochemical staining (IHC) images of tumors from nude mice stained with CD133 and MTDH. Histograms show the IHC score. Histograms show the IHC score (scale bars = 200 μm). ^aP < 0.01, ^bP < 0.001, ^cP < 0.0001. RNAi: Metadherin knockdown group; RNAi-NC: Metadherin knockdown control group.

A significant association was observed between immune cell infiltration and survival in patients with HCC. Using ssGSEA, we quantified immune cell infiltration scores in HCC samples from TCGA to understand the relationship between MTDH and infiltration. Firstly, the results demonstrated a significantly higher infiltration of activated B cell, immature B, memory B, activated CD8 T, gamma delta T, effector memory CD8 T, central memory CD8 T, and effector memory CD4 T cells in healthy tissue than in HCC. Similarly, higher infiltration of eosinophils, immature DCs, myeloid-derived suppressor cells, monocytes, macrophages, mast cells, monocytes, neutrophils, NK cells, NK T cells, and T helper (Th) cells were observed in healthy tissue than in HCC (Figure 7A). These findings suggest that immune cells are essential in the progression of HCC. Based on the median MTDH expression, we divided the liver cancer samples in TCGA into high and low MTDH expression groups and assessed immune cell infiltration in both groups. Notably, the levels of memory B (P < 0.0001), immature dendritic (P < 0.001), Th2 (P < 0.001), and central memory CD4⁺ T (P < 0.01) cells were higher in the high MTDH expression group than that in the low group. In the low MTDH expression group, activated CD8 T cells (P < 0.0001), macrophages (P < 0.01), activated B cells (P < 0.01), effector memory CD8 T cells (P < 0.01), mast cells (P < 0.01), eosinophils (P < 0.05), monocytes (P < 0.05), and Th1 cells (P < 0.05) were significantly increased (Figure 7B).

Open in New Tab   Full Size Figure   Download Figure

Figure 7 Infiltration of immune cells in TCGA samples using ssGSEA. A: Immune cell infiltration between LIHC samples and normal samples; B: Different immune cell infiltration patterns in high and low expression samples of Metadherin. ns: Not significant. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001, ^dP < 0.0001.

We examined the relationship between immune cells and MTDH expression (Figure 8A). Immature dendritic (r = 0.28, P < 0.0001; Figure 8B), Th2 (r = 0.26, P < 0.0001; Figure 8C), memory B (r = 0.25, P < 0.0001; Figure 8D), central memory CD4 T (r = 0.22, P < 0.0001; Figure 8E), central memory CD8 T, NK T, activated dendritic, activated CD4 T, and NK cells also showed a positive relationship with MTDH. Activated CD8 T cells (r = -0.23, P < 0.0001; Figure 8F), eosinophils (r = -0.13, P < 0.05; Figure 8G), activated B cells (r =-0.12, P < 0.05; Figure 8H), monocytes (r = -0.12, P < 0.05; Figure 8I), macrophages, and mast cells showed a negative correlation with MTDH.

Open in New Tab   Full Size Figure   Download Figure

Figure 8 Immune infiltration and Metadherin expression levels. A: Metadherin (MTDH) expression levels correlated with infiltration of immune cells; B: Correlations between MTDH and immature dendritic cells (r = 0.28, P < 0.0001); C: Correlations between MTDH and T helper 2 cells (r = 0.26, P < 0.0001); D: Correlations between MTDH and memory B cells (r = 0.25, P < 0.0001); E: Correlations between MTDH and central memory CD4 T cells (r = 0.22, P < 0.0001); F: Correlations between MTDH and activated CD8 T cell (r = -0.23, P < 0.0001); G: Correlations between MTDH and eosinophils (r = -0.13, P < 0.05); H: Correlations between MTDH and activated B cells (r =-0.12, P < 0.05); I: Correlations between MTDH and monocytes (r = -0.12, P < 0.05). MTDH: Metadherin.

ICIs (immune checkpoint inhibitors) are gaining increasing attention as tumor immunotherapy strategies for different types of cancers, facilitating improved prognosis in some patients. MTDH and the various immunosuppressive agents in the TISIDB database did not show significant correlation, as indicated by our online analysis (Figure 9A). However, in the analysis of correlation with immunostimulants (Figure 9B), a positive correlation was found between MTDH and MICB (r = 0.207, P = 5.76 × 10^-5; Figure 9C), NT5E (r = 0.201, P = 9.8 × 10^-5; Figure 9D), and TNFSF14 (r = 0.143, P = 0.00576; Figure 9E). Based on these results, MTDH may be involved in the regulation of tumor immunity.

Open in New Tab   Full Size Figure   Download Figure

Figure 9 Correlation between Metadherin expression and immunomodulators. A and B: Heat map showing the correlation between Metadherin (MTDH) and immunosuppressive agents and immunostimulants in hepatocellular carcinoma; C: Correlations between MTDH and MICB; D: Correlations between MTDH and NT5E; E: Correlations between MTDH and TNFSF14. MTDH: Metadherin.

Chemokines and their receptors induce cell migration. The expression of MTDH correlated with that of chemokines and receptors in immune cells, based on the data from the TISIDB database (Figure 10A and B). In HCC, MTDH expression positively correlated with CXCL2 (r = 0.224, P = 1.34 × 10^-5; Figure 10C) and negatively correlated with CX3CL1 (r = 0.245, P = 1.73 × 10^-6; Figure 10D) and CXCL12 (r = 0.208, P = 5.4 × 10^-5; Figure 10E). However, MTDH expression was not significantly correlated with chemokine receptor expression.

Open in New Tab   Full Size Figure   Download Figure

Figure 10  Chemokines and chemokine receptor correlates with Metadherin. A and B: A correlation analysis of Metadherin (MTDH) and chemokines and the receptors in LIHC is presented as a heat map; C: Correlations between MTDH and CXCL2; D: Correlations between MTDH and CX3CL1; E: Correlations between MTDH and CXCL12. MTDH: Metadherin.

Metadherin (MTDH) is a key oncogene in most cancer types, including hepatocellular carcinoma (HCC). Tumor stem cells are associated with tumorigenesis, metastasis, cell proliferation, and postoperative recurrence. The type and number of immune cells in the HCC microenvironment have prognostic value and can influence the response to immunotherapy. The impacts of MTDH expression on stem cell characteristics and immune cell infiltration in HCC remain unclear.

MTDH is a key oncogene in most cancers. It is important to explore the impact of MTDH on the prognosis of HCC patients and determine whether it affects tumor progression by influencing stem cell phenotype and immune infiltration.

This study aimed to investigate the effects of MTDH on tumor stemness and immunity in HCC.

Differential expression of MTDH in tissues was detected using TCGA and GEO databases, and immunohistochemistry was performed on HCC and para-cancerous tissue samples. MTDH was stably downregulated or overexpressed by lentiviral transfection in both HCC cell types. Invasiveness and migration were assessed using stromal infiltration and wound healing assays. HCC stem cells were obtained by culturing spheroids in a serum-free medium. Flow cytometry, immunofluorescence, and sphere formation assays were used to identify stem-like cells. Relevant gene expression was detected through western blotting and real-time quantitative reverse transcription-PCR. The effect of MTDH inhibition on tumor growth was investigated using in vivo tumor formation assays. Correlations between MTDH and immune cells, immunomodulators, and chemokines were analyzed using ssGSEA and the TISIDB database.

This study confirmed that the expression of MTDH in HCC tissues was higher than that in normal liver tissues and that the high expression of MTDH resulted in poor prognosis of patients with HCC. HCC cells overexpressing MTDH exhibited stronger invasion and migration abilities, exhibited a stem cell-like phenotype, and formed spheres, whereas MTDH inhibition attenuated these effects. MTDH inhibition suppressed tumor growth and CD133 expression in vivo. Correlation analysis showed that MTDH exhibited positive correlation with immature dendritic cells (DCs), T helper (Th)2 cells, central memory CD8⁺ T cells, memory B cells, activated DCs, natural killer (NK) T cells, NK cells, activated CD4⁺ T cells, and central memory CD4⁺ T cells. The results also showed that MTDH exhibited negative correlation with activated CD8⁺ T cells, eosinophils, activated B cells, monocytes, macrophages, and mast cells. Correlation analysis of MTDH expression with immunomodulators and chemokines showed that MTDH levels positively correlated with CXCL2 expression and negatively correlated with CX3CL1 and CXCL12 expression.

In HCC, MTDH expression is increased, leading to poor prognosis. MTDH promotes the acquisition of tumor cell stemness and tumor growth in HCC, influencing immune infiltration and immunotherapy.

Through database analysis, as well as in vivo and in vitro experiments, we confirmed that MTDH leads to poor prognosis in patients with HCC, promotes the acquisition of the tumor stem cell phenotype, and influences immune infiltration. The exact mechanism through which MTDH influences HCC stem cell and immune cell infiltration requires further exploration. This may provide a scientific basis for further understanding of the prognosis and treatment of patients with HCC.