Histone Deacetylase 1 Expression and Regulatory Network in Lung Adenocarcinoma Based on Data Mining and Implications for Targeted Treatment

Background and Aims Histone deacetylase 1 (HDAC1) codes a protein that is a component of the histone deacetylase complex. The abnormal expression of HDAC1 is strongly correlated with cell proliferation, differentiation, transcription, and translation. Through continuous screening of genes associated with changes in lung adenocarcinoma (LUAD), gene networks are formed to explore tumor pathogenesis and new therapeutic targets. Methods We evaluated HDAC1 gene survival analysis and its expression of LUAD using relevant websites and databases (TCGA and GEO databases). Through data mining, we determined the frequency and type of HDAC1 mutation, obtained the relevant heat map of the gene interaction network, completed the analysis of gene ontology and function enrichment, and understood the pharmaceutic of HDAC1. Results We found that HDAC1 expression was associated with the prognosis of patients with LUAD. In gene expression analysis, HDAC1 was highly expressed in LUAD, and the HDAC1 interaction gene network (MARCKSL, eIF3I) was closely related to cellular gene expression. Functional network analysis shows that the expression of HDAC1 is related to the monitoring point of the G1-S phase of the cell cycle and the activation of the Notch signaling pathway (CSL transcription factor), which is involved in the process of cell proliferation and differentiation and gene expression associated with new therapeutic targets. Conclusion Our data revealed the expression and potential regulatory factors of HDAC1 in LUAD of data mining, which laid a foundation for the study of the occurrence, development, and treatment of HDAC1 in LUAD.


Introduction
Lung adenocarcinoma (LUAD) belongs to non-small-cell carcinoma, and it is the fastest-growing lung cancer incidence rate at present [1,2]. Approximately 85% of lung cancer patients are diagnosed with non-small-cell lung cancer (NSCLC), which includes the histological subtypes of LUAD, squamous cell carcinoma, and large cell carcinoma. In the past, systemic cytotoxic chemotherapy has been the main treatment for advanced NSCLC, and new therapies are being developed. Currently, patients with LUAD are usually diagnosed using next-generation sequencing (NGS) for molecular testing to determine the best treatment for the patient. Epidermal growth factor receptor (EGFR) [3], BRAF-activated mutation, and anaplastic lymphoma kinase (ALK) [4] have been targeted as part of routine therapy [5][6][7]. At the same time, the success of immunotherapy has created a new paradigm of personalized therapy [8,9], extended the survival rate of oncogene-driven patients with advanced LUAD, and accelerated the development of new drugs for LUAD, but success remains limited. Te pathogenesis of LUAD is the interaction of polygenes and external factors, which is a complex pathological process of longterm development and formation. By continually screening genes for tumor-related changes to form gene networks, it may be possible to fnd out how tumors develop.
Histone deacetylase 1 (HDAC1) is known to be a component of the histone deacetylase complex. HDAC1 not only catalyzes histone acetylation and deacetylation in multisubunit complex but also interacts with retinoblastoma tumor suppressive proteins [10][11][12], which is a key factor in regulating eukaryotic gene expression control and cell proliferation and diferentiation. Recent studies have shown that HDAC1 is involved in the pathogenesis of many cancers. Related studies have shown that HDAC1 gene expression is highly expressed in LUAD tissues [13,14], and the related pathway afected by HDAC1 is closely related to the formation of cancer tissues.
At present, the related drugs targeting HDAC1 are mainly HDAC inhibitors [15,16]. Natural synthetic histone deacetylase inhibitors (HDACI) are compounds with different target specifcity and activity, mainly divided into four categories, including cyclic peptides, benzamide, short-chain fatty acids, and isohydroxamic acid [17]. HDACI induces diferent phenotypes in various transformed cells, including growth arrest, activation of apoptotic pathways, and autophagy death. All HDAC inhibitors can induce histone H3 hyperacetylation, which is related to the inhibition of proliferation, induction of cell diferentiation, and apoptosis. Meanwhile, normal tissue cells have relatively greater resistance to HDACI-induced cell death [18,19]. HDACI has been established as a new approach for the treatment of solid and hematologic tumors. By revealing the pathway of the role of HDAC1 in LUAD tissue, we may have the opportunity to fnd new targets and strategies for the diagnosis and treatment of LUAD, making the treatment of LUAD more accurate and efective.

Survival Analyses of HDAC1.
Sources of the Kaplan-Meier plotter database include Gene Expression Omnibus (GEO), Te European Genome-phenome Archive (EGA), and Te Cancer Genome Atlas (TCGA) [20]. Te tool enables univariate and multivariate Cox proportional hazard survival analyses using data generated from genomic, transcriptomic, proteomic, or metabolomic studies. We completed the survival analysis of HDAC1 in LUAD patients using this platform [21]. In statistical analysis, if the P value was less than 0.05, it was considered that the expression of HDAC1 was related to the prognosis of patients.

2.2.
Oncomine Analysis. Oncomine (https://www. oncomine.org/) is currently the world's largest oncogene chip database and integrated data-mining platform [22]. Te database collected data from 729 gene expression datasets and more than 90,000 cancers and normal tissue samples. Diferential expression analysis was performed by t-test as a measure of diferential expression, and the false discovery rate was calculated to correct for signifcance. HDAC1 copies the number of DNA in Oncomine calls to TCGA database data completion [23]. Te mRNA diferential expression analysis was performed on the data of the gene expression profle chip in the GEO database called by Oncomine. Te analysis studied a range of LUAD, including Weiss Lung, Okayama Lung, Hou Lung, and Beer Lung. We analyzed the expression of HDAC1 in LUAD tissues and normal tissues. If the fold change is less than 2, there is no signifcant diferential expression of the gene, and the t value in statistical analysis is less than 0.05, and it is considered that there is a diferential expression in LUAD tissues and normal tissues, and the lower the P value, the higher the degree of diference.
2.3. UALCAN Analysis. UALCAN (https://ualcan.path.uab. edu/index.html) is an efective cancer data online analysis and mining based on the TCGA database website [24]. Te website analyzes the relative gene expression of tumors, normal samples, and tumor subgroups using a t-test. In the process of the study, we used the UALCAN website to call the mRNA sequencing data in the TCGA database to carry out the comparative analysis of diferent ages, sexes, races, smoking or not, disease stages, tumor grades, and TP53 lesions with normal tissues and to verify the mRNA differential expression in the previous step. At the same time, the HDAC1 protein expression profle was analyzed, and the expression of HDAC1 gene protein at the total protein level, phosphorylated protein level, and pan-cancer level could be obtained. Trough Pearson correlation analysis on this website, we completed gene correlation analysis of HDAC1 and obtained gene sets interacting with target genes, to facilitate subsequent mRNA expression analysis of gene sets and molecular network analysis of interaction.

C-BioPortal
Analysis. C-BioPortal (https://www. cbioportal.org) is a powerful TCGA analysis platform that integrates the genomic data of 164 cancers such as TCGA, ICGC, and GEO [25,26]. We analyzed the mutation profle of the HDAC1 gene in LUAD samples to analyze mutation levels to analyze how the gene functions in cancerous tissues.

LinkedOmics Analysis.
LinkedOmics is a publicly available portal that includes multiomics data from 32 TCGA cancer types [27]. It also includes MS-based proteomics data generated by the Clinical Proteomic Tumor Analysis Consortium (CPTAC). Tus, heat maps of mRNA expression of some genes are positively associated with HDAC1, and those negatively associated with HDAC1 in LUAD were obtained.

WEBGESTAIT Analysis.
WEBGESTAIT enrichment analysis (https://www.webgestalt.org/option.php) is a focus of the online website [29,30], supports multiple enrichment analysis methods, and covers the functional annotation database comprehensively. We used this website to obtain the enrichment results of HDAC1 and its related gene sets, which intuitively showed that genes were related to those pathways in Gene Ontology (GO), to explore and verify the functions of genes.

KOBAS (KEGG Orthology Based Annotation System)
Analysis. KOBAS 3.0 (https://kobas.cbi.pku.edu.cn/kobas3) is widely used in gene and protein function annotation databases [31]. We analyzed the Kyoto Encyclopedia of Genes and Genomes (KEGG) via the website. After entering related search options, we found relevant KEGG functional pathways. We investigated the possible role of HDAC1 in cancer tissue formation.
2.9. DRUGSURV Analysis. DRUGSURV (https://www. bioprofling.de/GEO/DRUGSURV/) is a tumor drug database for querying and analyzing genes related to clinical prognosis [32,33]. Te database collects about 1700 FDAapproved drugs. Te database contains 5000 investigational drugs and related target genes, as well as 17 associated tumors and about 50 clinical prognostic data. Trough this website, we can check the latest progress of HDAC1 drug development and related target genes.

Gene Survival Analysis of HDAC1.
We obtained the survival analysis of HDAC1 for LUAD patients using the Kaplan-Meier plotter platform ( Figure 1). Te results showed that there was a statistically signifcant diference in survival between patients with high HDAC1 expression and patients with low HDAC1 expression (hazard rate: HR � 1.2, log-rankP � 0.0056). Tis suggests that high expression of this gene is a risk factor leading to increased mortality.

Expression of HDAC1 in LUAD.
We analyzed and evaluated the copy number variation (CNV) analysis of DNA of the target gene HDAC1 and the chip diference analysis of HDAC1 mRNA gene expression profle through TCGA and GEO databases (Figures 2(a)-2(f )). Tere was no signifcant diference in gene copy number in cancer tissues compared with normal tissues (P > 0.05). HDAC1 mRNA expression in some LUAD tissues was signifcantly higher than that in normal tissues (P < 0.01). Although the data showed that the diferential multiples of HDAC1 were all within 2, HDAC1 mRNA overexpression ranked in the top 2%-6% in some cancer tissues. Meanwhile, we further understood the mRNA expression of HDAC1 through the mRNA sequencing data in the TCGA database which can be called by the UALCAN database (Figures 3(a)-3(f )). We see, according to age, gender, race, smoking, disease stage, tumor grade, and TP53 lesions, whether the database HDAC1 in cancerous tissue translation level is signifcantly higher than that of normal tissue (Figures 4(a)-4(c)). HDAC1 gene expression increases sharply after TP53 mutation, and deacetylation of HDAC1 gene expression product may lead to the occurrence of cancer. At the same time, diferential analysis of HDAC1 protein expression profle was performed to analyze the level of total protein and phosphorylated protein, as well as the expression of HDAC1 gene protein at the pan-carcinoma level ( Figures 5-7). In the protein expression profle, HDAC1 expression in cancer tissues was higher than that in normal tissues in most of the protein expression categories. It can be seen that the mRNA expression of HDAC1 may be closely related to the pathogenesis of LUAD. Perhaps, we can explore the correlation between HDAC1 and LUAD as a diagnostic indicator or treatment plan.

Genomic Changes of HDAC1 in LUAD.
We obtained the frequency and type of HDAC1 mutations through the cBioPortal database (Figure 8(a)). Te results showed that 18 of 1833 individuals had HDAC1 mutations (1%), including missense mutations, deletion mutations, deep deletions, and amplifcations. Te most common type of DNA variation in HDAC1 is amplifcation.

Biological Interaction Gene Network of HDAC1 in LUAD.
We obtained some genes that may interact with HDAC1 by TCGA analysis in UALCAN. By Pearson correlation (Person

Discussion
Histone acetylation and deacetylation play an important role in the regulation of gene expression in eukaryotic cells [34,35]. Te protein encoded by HDAC1 is part of the histone deacetylase complex. Current drug progresses related to HDAC1 show that HDACi can increase the acetylation degree of intracellular histones and nonhistones, improve the expression level of P21, P53, and other genes, and then achieve the efect of inhibiting tumor cells [36,37]. Tis kind of drug can also show a good synergistic therapeutic efect when combined with a variety of chemotherapy drugs and has low toxicity in the efective inhibitory dose range, which has made HDACi a new targeted antitumor drug with wide application. Te discovery of new marker targets to establish related gene networks will help diagnose and improve treatment efects.
In previous work, we found that high expression of HDAC1 is a risk factor for LUAD patients, leading to increased mortality. We analyzed transcription and sequencing data from more than 1000 clinical samples from      Journal of Oncology 9 GEO and TCGA databases, confrming that the HDAC1 gene had little diference in copy number in LUAD and signifcantly higher mRNA and protein expression levels than in normal lung glandular tissue. HDAC1 ranks 2-6% among the genes upregulated in some tissues of LUAD. In the LUAD patient sequencing database in TCGA, the mutation frequency of HDAC1 is 1%. Te mutation type is mainly amplifcation. 18 of 1833 people have mutations in HDAC1. We speculate that there are fewer mutations in the genetic content of chromosome HDAC1 in LUAD tissues, but the highly expressed promoter in cancer tissues participates in assisting the expression of HDAC1, thereby increasing the efciency of transcription or translation.
In the analysis of HDAC1-related gene network interaction diagram and coexpression diagram, related person diagram, positively correlated gene set heatmap, and negatively correlated gene set heatmap, it can be seen that HDAC1 is positively correlated with MARCKSL1, EIF3I, and other genes. Among them, the MARCKSL1 is related to the occurrence and development of many types of cancers. Relevant literature has proved that MARCKSL1 promotes the progression of LUAD by regulating epithelialmesenchymal transition (EMT) and can be used as a new therapeutic target for LUAD [38]. EIF3I is a proto-oncogene that is overexpressed in a variety of tumors. It upregulates vascular endothelial growth factor A, promotes cell proliferation, and promotes embryonic development and angiogenesis in tumorigenesis [39]. EIF3I is very important for the regulation of protein synthesis, cell proliferation, cell cycle progression, and tumorigenesis [40,41]. Our results show that HDAC1 has a strong correlation with overexpressed genes in tumors.
In the enrichment analysis of the target gene set, the gene set related to HDAC1 is enriched in multiple GO pathways. Our enrichment analysis results show that the expression of HDAC1 protein is mainly located in the nucleus, its molecular function is mainly protein binding, and the biological process involved is mainly biological regulation. HDAC1 also participates in many molecular signaling pathways on the KEGG pathway. We found that HDAC1 in LUAD is related to the kinase network including CDK and cyclin and the Notch signaling pathway [42,43]. Tese kinases are protein complexes that mainly regulate the cell cycle and are closely related to the stable proliferation of cells. Overexpression of HDAC1 can cause abnormal cell cycle regulator E2F transcription factor, which leads to the production of cancer cells [44]. Te E2F transcription factor is one of the key regulators of the G1-S phase transition in the cell cycle. Our analysis shows that the E2F transcription factor is an important target of HDAC1, and HDAC1 regulates the proliferation of LUAD cells through the abnormality of this factor.
Regarding the Notch signaling pathway [45], it has been documented that the overexpression of intracellular segments (NICD) in mouse alveolar epithelial cells induces hyperplasia and eventually leads to LUAD. Relevant literature proves that Notch1 is necessary for Kras-induced LUAD and can control tumor cell survival through TP53 [46]. Notch1 and Notch3 signal transduction promotes tumor cell proliferation and inhibits apoptosis in certain NSCLC cell lines [47]. In our results, the overexpression of HDAC1 can lead to the inhibition of the CSL transcription factor, thereby inhibiting the transcription of related genes, inhibiting the activation of the Notch signaling pathway, and causing the Notch signaling pathway to be abnormal. Te target genes of the Notch signal are mostly basic helix-loophelix transcription factors, such as nuclear factor-κB (NF-κB), cyclin D1, c-myc gene, P21 gene, P27 gene, AKT serine/ threonine kinase, and the mechanistic target of rapamycin (mTOR). Te target genes of the Notch signaling pathway  have been well documented in the occurrence and progression of tumors. Tey regulate the transcription of other genes directly related to cell diferentiation. In the relevant literature, abnormal regulation of the Notch pathway may occur through a variety of mechanisms, including mutation activation or inactivation, overexpression, posttranslational modifcation, and epigenetic regulation. More and more evidence shows that Notch1 is a putative oncogene in LUAD.
In human non-small-cell lung cancer, the gain-offunction mutation of the Notch1 gene and the weakening of the pathway due to the loss of NUMB have been described. Relevant literature shows that the Notch signaling pathway has a cancer-promoting efect on lung cancer. Te Notch signaling pathway interacts with multiple signaling pathways, including Wnt, TGF-β, and HER-2. We suspect that the overexpression of HDAC1 causes the transcriptional inhibition of Notch signal target genes to inhibit the progression of LUAD. Tis study provides relevant evidence for the correlation of HDAC1 overexpression in the occurrence and progression of LUAD and its potential as a marker for LUAD. Tey include HDAC1-interacting genes such as Positively Correlated Significant Negatively Correlated Significant Genes Z-Score Group Te left side of each graph is a positive correlation, and the right side is a negative correlation. In the graph, the darker the red is, the higher the correlation, while the darker the blue is, the lower the correlation. Journal of Oncology 13 EIF3I, MARCKSL1, and kinase network genes including cyclin-dependent kinase (CDK). In addition, the Cyclin and Notch signaling pathways involved in HDAC1 are closely related to cell growth, development, diferentiation, and difusion. HDAC1 is particularly related to several tumorrelated kinases (such as CDK), transcription factors (EIF3I, CSL, E2F), and signaling pathways (NOTCH).
In this study, we used online tools based on the most popular bioinformatics theory to analyze target genes from the tumor data in the public database. Compared with traditional chip screening, this method has the advantages of a large sample size, low cost, and simple operation. At the same time, the TCGA database also has some limitations. First, there is no sufcient data on the LUAD sample in TCGA. Only three ethnic groups were included in the LUAD sample, and stage 4 patients were relatively rare. Te absence and insufciency of data directly afect our analysis. Another limitation is that transcriptome sequencing can only detect static mutations; it cannot directly provide information about the level of protein activity or expression. Tese issues should be addressed in subsequent activities using molecular biology techniques. Our results suggest that HDAC1 may be used as a prognostic marker for cancer treatment in future clinical practice. At the same time, due to its important role in tumor genesis and development, this study provides a theoretical basis for the HDAC1 target therapy of LUAD.

Conclusion
With the understanding of the related role of HDAC1 in cells, we concluded from the results of HDAC1 expression in LUAD, mutation frequency and form, gene interaction, and functional enrichment analysis that the abnormal transcription factors and related signal pathway changes caused by HDAC1 overexpression are related to the occurrence and development of LUAD. Tere are limitations in the process of our study, which need to be improved through further biochemical experiments in subsequent activities. Our study helps uncover gene networks associated with LUAD and new therapeutic targets.

Data Availability
All data generated or analyzed during this study are included in this published article.

Ethical Approval
Te work was approved by the Guangdong Medical University Ethics committee and by the Declaration of Helsinki of the World Medical Association.

Consent
Informed consent forms are not required for patient data extracted from databases.

Conflicts of Interest
Te authors declare that they have no conficts of interest.