ADAM12 is an independent predictor of poor prognosis in liver cancer

Disintegrin and metalloproteinase 12 (ADAM12) is thought to trigger the occurrence and development of numerous tumours, including colorectal, breast, and pancreatic cancers. On the basis of The Cancer Genome Atlas (TCGA) datasets, in this study, the relationship between ADAM12 gene expression and hepatocellular carcinoma (HCC), the prognostic value of this relationship, and the potential mechanisms influencing HCC development were evaluated. The results showed that the ADAM12 gene was significantly and highly expressed in liver cancer tissue. The high expression of the ADAM12 gene in liver cancer tissue significantly and positively correlated with T stage, pathological stage, and residual tumour. Kaplan–Meier and Cox regression analyses revealed that ADAM12 gene expression is an independent risk factor influencing the prognosis of patients with liver cancer. Pathway analyses of ADAM12 in HCC revealed ADAM12-correlated signalling pathways, and the expression level of ADAM12 was associated with immune cell infiltration. In vitro experiments demonstrated that the expression level of ADAM12 in Huh-7 and Hep3B cells was significantly higher than that in other HCC cells. ShRNA transfection experiments confirmed that the expression levels of TGF-β and Notch pathway-related proteins were significantly decreased. An EdU cell proliferation assay showed that a low level of ADAM12 gene expression significantly inhibited the proliferative activity of HCC cells. Cell cycle experiments showed that low ADAM12 expression blocked the G1/S phase transition. Overall, this research revealed that high ADAM12 gene expression implies a poor prognosis for patients with primary liver cancer. In addition, it is a potential indicator for the diagnosis of liver cancer.

Scientific Reports | (2022) 12:6634 | https://doi.org/10.1038/s41598-022-10608-y www.nature.com/scientificreports/ have displayed better therapeutic effects and weaker side effects in advanced hepatocellular carcinoma than traditional therapies 6,7 . Disintegrin and metalloproteinases (ADAMs) are transmembrane glycoproteins anchored to the cell membrane that belong to the Metzincin superfamily of Zn 2+ -dependent metalloproteinase enzymes [8][9][10] . To date, as many as 40 members of the ADAM protein family have been identified, and they have been shown to regulate cell phenotypes through their effects on cell adhesion, migration, proteolysis, and signal transduction [10][11][12] . In recent years, several studies have shown that this family promotes the occurrence and development of tumours by targeting cell development, regulating the inflammatory response, and releasing membrane-bound proteins 13,14 .
The ADAM12 gene is generated by ADAM splicing; is primarily found in the bone and cartilage; and is expressed in the brain, liver, heart, and muscles [15][16][17] . The human ADAM12 gene is located on chromosome 10q26. 3, and in tumour cell lines, it is expressed in two different spliced forms 16 . Additionally, several studies have shown that the expression of the ADAM12 gene is upregulated in pancreatic, colorectal, gastric, lung, and breast cancers and leads to poor prognosis [17][18][19][20][21] . Notably, the ADAM12 gene is a binding partner of the activated protease C receptor during liver fibrosis, and both ADAM12 and protease C are highly expressed in liver cancer and liver fibrosis 22 . Liver fibrosis and cirrhosis caused by chronic hepatitis B are important risk factors for liver cancer 23,24 ; therefore, ADAM12 is closely related to liver injury and liver cancer. Nonetheless, reports on ADAM12 in primary liver cancer remain rare. Although studies have suggested that the ADAM12 gene may be related to the invasion and progression of liver cancer cells 25 , the potential mechanisms and clinical relevance of these actions remain to be clarified.
In this study, we evaluated the contribution of ADAM12 gene expression to the prognosis of liver cancer using liver cancer samples obtained from the TCGA database. Moreover, we clarified the biological pathway and mechanism of the ADAM12 gene implicated in the regulation of the occurrence and development of liver cancer via gene enrichment analysis (GSEA) and immune cell invasion analysis. Our findings will help in the discovery of novel prognostic biomarkers and prediction of the potential molecular mechanisms influencing the prognosis of liver cancer.

Results
Clinical features of patients with hepatocellular carcinoma. The clinical data used to characterize 374 patients with hepatocellular carcinoma were downloaded from the TCGA database and divided into high and low groups based on the median ADAM12 gene expression level. Comprehensive information on these clinical data is provided in Table 1. ADAM12 is highly expressed in liver cancer tissues. On the basis of TCGA data, the expression differences in ADAM12 gene mRNA in different tumour tissues and normal tissues were analysed by Wilcoxon signed-rank test (Fig. 1A). The results revealed that the ADAM12 gene was expressed in various tumours. In hepatocellular carcinoma (LIHC) and cholangiocarcinoma (CHOL), significant differences were noted between tumour tissue and normal tissue. In hepatocellular carcinoma, the expression of the ADAM12 gene in tumour tissues was significantly higher than that in normal tissues (p = 6.4e−06) (Fig. 1B). Subsequently, Wilcoxon signed-rank test was performed to analyse the paired sample data of ADAM12 expression in 50 cases of liver cancer and adjacent tissues. The results indicated that the expression of ADAM12 in normal tissues was significantly lower than that in cancer tissues (p = 5.7e−05) (Fig. 1C).
Relationship between ADAM12 gene expression and clinicopathologic features in patients with primary liver cancer. The Wilcoxon signed-rank test and logistic regression analysis were performed to evaluate the relationship between ADAM12 gene expression and clinicopathological variables in patients with hepatocellular carcinoma. The expression of the ADAM12 gene in hepatocellular carcinoma was significantly correlated with T stage (p = 0.01), age (p = 0.03), sex (p = 0.02), pathological stage (p = 7.2e−03), and histological grade (p = 0.01) (Fig. 2). Thereafter, the relationship between ADAM12 gene expression and the clinicopathological characteristics of the HCC patients was analysed through univariate logistic regression. The results showed that high expression of the ADAM12 gene was significantly correlated with T stage, clinicopathological stage, sex, and ethnicity; nevertheless, no significant difference was found between ADAM12 gene expression and clinicopathological variables, including N stage, M stage, histological grade, vascular invasion (Table 2).
Clinical significance of ADAM12 gene expression in the prognosis of liver cancer. Furthermore, the Kaplan-Meier Plotter database was used to analyse the clinical significance of ADAM12 gene expression in the prognosis of liver cancer. The results showed that the overall survival time of the group with high ADAM12 gene expression was shorter than that of the group with low ADAM12 gene expression (p = 4e−05) (Fig. 3A). The mean OS time in the low ADAM12 expression group at 12 (1 year), 36 (3 years), and 60 months (5 years) was significantly longer than that in the high expression group (Fig. 3B-D). A univariate Cox regression analysis revealed that ADAM12 gene expression was a high-risk factor for HCC (HR, 1.818; CI, 1.280-2.582; p < 0.001) ( Table 3). A multivariate Cox regression analysis showed that high ADAM12 gene expression was an independent prognostic factor related to OS (HR, 1.552; CI, 1.054-2.285; p = 0.026) (Fig. 3E).
GSEA of the ADAM12 gene. To explore the possible mechanism of ADAM12 gene effects in liver cancer, data from the TCGA database were utilized to perform gene set enrichment analysis (GSEA

Relationship between ADAM12 gene expression and immune cell infiltration.
To comprehensively investigate the role of ADAM12 in HCC, we selected the immune infiltrating algorithm (ssGSEA) and Spearman correlation to analyse the association between ADAM12 expression levels and subsets of infiltrating immune cells. The expression of the ADAM12 gene was found to be positively correlated with macrophages (p < 0.01), immature dendritic cells (p < 0.01) and follicular T cells (p < 0.01), but negatively correlated with helper T cells 17 (p < 0.01) (Fig. 5). In addition, ADAM12 was associated with markers of macrophages, immature dendritic cells, follicular T cells and helper T cells (Table 5).
Differential expression of ADAM12 in hepatocellular carcinoma tissues and hepatocellular carcinoma cells. To verify the bioinformatics analysis results and determine the correlation between ADAM12 www.nature.com/scientificreports/ gene expression and liver cancer, we first compared the expression of ADAM12 in cancer tissues and adjacent tissues taken from liver cancer patients. The results showed that the expression level of the ADAM12 protein in the cancer tissues was significantly higher than that in the adjacent tissues (Fig. 6A, p < 0.05). Next, we detected the protein expression level of ADAM12 in different HCC cell lines. A Western blot analysis indicated that the  www.nature.com/scientificreports/ ADAM12 protein expression level was relatively high in Huh-7 cells and Hep3B cells (Fig. 6B, p < 0.05). In addition, we validated shRNA interference sequences. The results showed that the third interference sequence tested was effective in inhibiting ADAM12 expression (Fig. 6C, p < 0.05).

Knocking down ADAM12 expression inhibited the proliferation of HCC cells and blocked the G1/S transition.
To further clarify the role of ADAM12 expression in the progression of HCC, we performed related functional experiments. An EdU proliferation experiment was performed to determine whether the expression level of ADAM12 affected the cell proliferation rate. The results showed that inhibition of ADAM12 expression resulted in a decreased cell proliferation rate compared with that of the control group (Fig. 7A, B). To further explore the specific reasons that the ADAM12 gene affects the proliferation and viability of liver cancer cells, we carried out cell cycle experiments. The results showed that low ADAM12 gene expression blocked the  www.nature.com/scientificreports/ transition of liver cancer cells from the G1 phase to the S phase (Fig. 7C, D). CyclinD1 is a key protein affecting the G1 phase transition 29 . Our results showed that the expression level of cyclin D1 decreased after ADAM12 expression was reduced (Fig. 7E, F). These results suggested that low expression of ADAM12 may inhibit the proliferation of hepatoma cells by suppressing cell cycle progression, further suggesting that the ADAM12 gene is related to poor prognosis in HCC. Fig. 4, the GSEA revealed that in the ADAM12 gene highexpression group, the most significantly enriched pathway was the Notch signalling pathway. To confirm this finding, specific ADAM12 shRNA was first transfected into cells to inhibit ADAM12 expression. The results showed that the expression levels of key Notch signalling pathway proteins Notch2, Hes1 and Jagged1 were significantly downregulated in the ADAM12 low expression group compared with the control group, suggesting that the Notch pathway was blocked (Fig. 8A, B). Similarly, the expression level of TGF-β was also decreased upon ADAM12 gene expression inhibition (Fig. 8A, B). The γ-secretase inhibitor DAPT is a potent inhibitor of the Notch signalling pathway, and therefore, it was added to the cell cultures; however, inhibition of Notch signalling did not change the expression of ADAM12 (Fig. 8C), but it inhibited the proliferation of liver cancer cells (Fig. 8D). Taken together, these data indicated that the ADAM12 gene may be involved in the occurrence and development of HCC through its effects on the Notch signalling pathway. The in vitro experimental results were consistent with those of the bioinformatics analyses.

Discussion
Primary liver cancer is caused by various complex factors; however, it primarily results from increased liver cell damage. Liver fibrosis is a chronic pathological process caused by excessive deposition of ECM, which induces liver injury and subsequent changes in the microenvironment to induce primary liver cancer occurrence and development 30,31 . Considering the role of liver fibrosis in liver cancer, Elsharkawy's team proposed the "inflammation-fibrosis-cancer axis" theory in 2007 32 . Notably, the relationship between ADAM12 expression and liver cancer has rarely been reported; however, ADAM12 has been shown to promote liver fibrosis. Studies have indicated that ADAM12 is negligibly expressed in normal liver tissue, but after it is highly expressed, ADAM12 activates hepatic stellate cells by activating transforming growth factor-β (TGF-β) 25 . Moreover, Nathalie's team proposed that RACK1 is a chaperone for ADAM12 promotion of liver fibrosis 22 . Therefore, the effect of ADAM12 expression on hepatic fibrosis implies that ADAM12 might be closely related to the occurrence and development of liver cancer. We analysed the integrated data and discovered that the expression level of the ADAM12 gene in primary liver cancer tissues was significantly higher than that in paracancerous tissues. The survival analysis and logistic regression analysis revealed that patients with high ADAM12 gene expression exhibited a low survival rate and poor prognosis. The multivariate Cox regression analysis showed that the ADAM12 gene is an independent risk factor for HCC. Tumour staging (TNM) is used to evaluate the number and location of malignant tumours in vivo, and it is used to guide clinical treatment, to a certain extent, since each stage (I to IV) is associated with different prognostic characteristics 33,34 . In the present study, the expression level of the ADAM12 gene was significantly and positively correlated with tumour size and pathological stage, indicating that the ADAM12 gene can potentially be used as an indicator of liver cancer stage.
Excessive cell proliferation and rapid metastasis are typical features of the sustained development of malignant tumours; these processes are regulated by multiple signalling pathways in vivo [35][36][37] . We found that highly expressed ADAM12 was primarily enriched in the Notch, GnRH, Hedgehog, TGF-β, JAK/STAT, MAPK, Calcium, Neurotrophin, and Fc Epsilon signalling pathways. Previous studies have implicated Notch in the proliferation, migration, and metastasis of various cancer cells [38][39][40] . However, Notch signalling may play a contradictory role due to the interaction between pathway regulatory mechanisms and the microenvironment 41,42 . However, several lines of evidence have indicated that Notch signalling pathways are significantly associated with cirrhosis,  [43][44][45] . Similar to the Notch pathway, the TGF-β signalling pathway plays a dual role in the development of HCC and inhibits the early development of HCC 46 . Nonetheless, hepatoma carcinoma cells undergo the EMT after responding to TGF-β, and therefore, the migration and invasion of liver cancer cells are increased 47,48 . Importantly, our study validated the association of ADAM12 gene expression with the Notch and TGF-β signalling pathways. Viral hepatitis and NASH are important causes of liver cancer development. Liver fibrosis due to viral hepatitis and NASH has been shown to be a risk factor for the development of hepatocellular carcinoma [49][50][51] . It has been suggested that ADAM12 is significantly associated with liver fibrosis 52 . The results of this study showed that ADAM12 was positively correlated with the expression of TGF-β (a marker of fibrosis) in liver cancer. In addition, activation of the Notch pathway increases the rate of liver fibrosis 53 . Therefore, the expression of the ADAM12 gene in the liver may promote the progression of hepatocellular carcinoma through the formation of liver fibrosis, but the data are not conclusive. In contrast to the complex roles of the Notch and TGF-β signalling pathways, the Hedgehog signalling pathway has been found to be consistently abnormally activated in association with tumour progression, metastasis, and drug resistance [54][55][56] . In HCC, the high expression of SMO and GLI1, members of the Hedgehog signalling pathway, directly triggers the formation of larger tumours     57,58 . In addition, the JAK/STAT and MAPK signalling pathways are associated with antitumour immunity 59 . Both the JAK/STAT and MAPK pathways have been confirmed to be crucial targets in related studies of HCC resistance 60,61 .
In recent years, tumour immune evasion has been a hot topic in research related to tumour treatment. Several immune cells, including macrophages, T cells, and autonomous NK cells reside in the microenvironment of tumour tissues. These cells directly or indirectly affect the tumour cell microenvironment and regulate tumour cell behaviour. For instance, in hepatocellular carcinoma, the antigen-specificity of invasive T cells is highly correlated with tumour control that is specifically manifested as CD8 + T cells, which are associated with an effective antitumour response 62 . Notably, by upregulating IL-10 expression, HIG2 facilitates HCC evade the killing induced by NK cells, promoting HCC cell recurrence and metastasis 63 . Because of these outcomes, immune cell therapy in hepatocellular carcinoma has broad application prospects. The expression of ADAM12 may be associated with immune cell infiltration in hepatocellular carcinoma; these invading cells include macrophages,

Conclusion
In conclusion, we used bioinformatics to explore the close correlation between high ADAM12 gene expression and hepatocellular carcinoma. In addition, we evaluated the feasibility of ADAM12 gene expression as a prognostic factor for hepatocellular carcinoma and predicted the possible ways that ADAM12 affects the progression of liver cancer, thereby providing a reliable basis for the prevention and prognosis of hepatocellular carcinoma. Finally, inhibition of ADAM12 gene expression significantly blocked the G1 phase transition and thus reduced cell proliferation in our experiments. Furthermore, knocking down ADAM12 expression inhibited Notch pathway activation and reduced the TGF-β expression level. Although Huh-7 and Hep3B cell lines were mainly used in the experiments, we will verify the results with human and murine HCC tissues in future studies. In summary, www.nature.com/scientificreports/ we speculate that ADAM12 exhibits great potential for promoting HCC progression, leading to a poor prognosis for patients with HCC.

Methods
Data mining the TCGA database. The gene expression data and basic clinical characteristic data, based on 50 samples of normal liver tissue and 374 samples of liver cancer tissue (workflow type, HTSeq-FPKM), were downloaded from the TCGA website (https:// portal. gdc. cancer. gov/ repos itory), which is a publicly available database. The expression information of the ADAM12 gene was obtained through high-throughput sequencing of TCGA data. First, an analysis of differential ADAM12 gene expression was performed, and the results are presented with a box diagram and paired difference diagram. Ultimately, the RNA-Seq HTSeq-Counts gene expression data for liver cancer patients and the clinical data were analysed with R software (version 3.6.3).
Gene set enrichment analysis (GSEA). GSEA software was downloaded from the GSEA website (GSEA V4.1.0; https:// www. gsea-msigdb. org/ gsea/ index. jsp). Based on the median ADAM12 gene expression level, the patient data were categorized into high and low groups. Then, these TCGA data were prepared in text format and imported into GSEA software. Notably, the gene set arrangements were repeated 1,000 times for each analysis. For the GSEA, the P value, FDR value, ES value, and NES value were mainly assessed; these are widely recognized screening parameters. Gene enrichment sequencing was first performed based on the NES values; then, in the gene enrichment analysis, the genes with expression differences that met the p < 0.05 and FDR < 25% criteria were considered to be significantly differentially expressed 64 .
Survival analysis. The Kaplan-Meier Plotter (http:// kmplot. com/ analy sis/) was used to analyse the prognosis of 25 types of tumours. The ADAM12 gene in hepatocellular carcinoma was analysed using this website, and overall survival was estimated by calculating logarithmic grade P values and hazard ratios (HRs).  EdU assay 65 . The proliferation ability of cells was detected with an EdU assay kit following the protocol provided by the manufacturer. In brief, cells were seeded into 24-well plates. After adherence, the cells were incubated with EdU solution for 2 h. Following three washes with PBS, the cells were fixed with 4% paraformaldehyde, neutralized with glycine, and reacted with Apollo® fluorescence dyes. Finally, Hoechst 33,342 reaction solution was used to stain nuclei. The cells were observed by inverted fluorescence microscopy (Olympus microscope, Tokyo, Japan) at 100 × magnification. The proportion of EdU-positive cells (red fluorescence) to Hoechststained cells (blue fluorescence) was calculated. The results were analysed using ImageJ version 1.52 software.
Western blot analysis. The cells were dissolved in RIPA lysis buffer and PMSF (RIPA:PMSF = 100:1, v/v), and cell lysates were collected and then centrifuged for 10 min at 12,500 g. The protein concentration was measured with an ND2000 microspectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Proteins were separated by SDS-PAGE (10% gels), transferred to PVDF membranes, blocked with 5% free fat milk at room temperature for 2 h and incubated with primary antibody overnight in a 4° refrigerator. The next day, the membranes were incubated with the corresponding secondary antibody at room temperature for 2 h. ECL (chemiluminescence) fluid and an imaging system were used for exposure and data analysis. Protein expression levels were semiquantified by ImageJ version 1. Immunohistochemistry (IHC). Eight pairs of clinical liver cancer samples were collected, and paraffin sections were obtained from the Department of Pathology, Yijishan Hospital, Wannan Medical College. The expression of ADAM12 in the liver cancer tissues and adjacent tissues was detected by IHC. First, xylene was used to deparaffinize and a diluted alcohol gradient was used for hydration. Tissue sections were treated for antigen retrieval with citrate antigen retrieval buffer. Then, the tissue was placed in 3% hydrogen peroxide to block endogenous peroxidase and blocked with serum for 30 min. The tissue was incubated with anti-ADAM12 antibody overnight at 4 °C. The next day, labelling with the corresponding secondary antibody was performed, followed by DAB colour development. Finally, the stained tissue was visualized with a light microscope.
Cell cycle analysis. Cells were collected according to kit instructions (China KGI, Cat: KGA511) and immobilized overnight with a final concentration of 70% ethanol. Next, the cells were washed three times with PBS and centrifuged to remove the supernatant. Finally, the cells were stained according to the instructions of a cell cycle assay kit and maintained in the dark for 30 min at room temperature. A flow cytometer (BD, USA) was used to detect the cell cycle distribution.
Statistical analysis. R software (version Rx64 V3.6.3) was used for statistical analysis. Wilcoxon rank-sum test was performed to analyse the difference in the ADAM12 gene expression between the normal liver tissue and liver cancer tissue. In addition, Wilcoxon signed-rank test, logistic regression, and Kruskal-Wallis tests were performed to analyse the relationship between ADAM12 gene expression and clinicopathological features, excluding patients with incomplete clinical data. The overall survival curve was plotted on the basis of a Kaplan-Meier analysis, and univariate and multivariate Cox regression analyses were performed to assess the effects of the ADAM12 gene and clinical characteristics on overall survival. Experiments were performed in triplicate, and the data are shown as the means ± SD. The statistical significance of differences was determined by Student's t test for comparisons of two groups. p < 0.05 was considered to be statistically significant.
Ethics approval. This study was approved by the Ethics Committee of Scientific Research and New Technology of Yijishan Hospital of Wannan Medical College with the approval number of 202095, and obtained informed consent from all patients who participated in this study. All methods were performed in accordance with the Helsinki declaration guidelines and regulations.
Consent for publication. We agree to publish the manuscript.
Scientific Reports | (2022) 12:6634 | https://doi.org/10.1038/s41598-022-10608-y www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.