Glioma is the most common tumor of the central nervous system (CNS). It accounts for 30% of all CNS tumors and 80% of all malignant brain tumors (Ostrom et al., 2019). Progress has been made in glioblastoma (GBM) treatment by combining maximal surgical resection with radiotherapy and concurrent and adjuvant temozolomide chemotherapy. Nevertheless, the two- and five-year survival rates are only 25 and 10%, respectively, (Stupp et al., 2005; Stupp et al., 2009). Tumor heterogeneity is a major challenge in precise glioma diagnosis and therapy (Neftel et al., 2019). Hence, molecular signatures for glioma are urgently required (Song et al., 2014).

Growing evidence suggests that epigenetic and genetic changes are pivotal in malignant disease onset and progression (Jones and Baylin, 2007). Histone deacetylases (HDACs) are important regulators of gene expression and control a broad array of physiological processes such as differentiation, apoptosis, survival, proliferation, and autophagy. They are also involved in cancer pathogenesis (Glozak and Seto, 2007; Wilson et al., 2008; Kang et al., 2014). There are four HDAC families and 18 individual HDACs. The zinc-dependent HDAC family comprises 11 isoforms divided into class I (HDAC1–3 and 8), class IIa (HDAC4, 5, 7, and 9), class IIb (HDAC6 and 10), and class IV (HDAC11) (Kashyap and Kakkar, 2020). Class III HDACs are homologous to yeast Sir2 and participate in transcriptional silencing. However, they have a deoxyhypusine synthase-like NAD/FAD-binding domain that is distinct from those of the other HDAC classes (Park et al., 2021). HDAC1 and two act mainly via nucleosome remodeling and deacetylase, switch independent 3, mitotic deacetylase, and corepressor of REST complexes. HDAC3 is recruited only by the nuclear receptor corepressor complex (Millard et al., 2017). HDAC6 regulates Hsp90, tau, and the cytoskeleton by interacting with tubulin and cortactin. It recognizes ubiquitinated proteins and induces aggregate formation (Hubbert et al., 2002; Kovacs et al., 2005; Dompierre et al., 2007; Zhang et al., 2007; Cohen et al., 2011; Simões-Pires et al., 2013). HDAC10 is a polyamine deacetylase (Ho et al., 2020), and HDAC11 has limited homology to class I and II enzymes (Li et al., 2015).

HDACs play important roles in many diseases. HDAC1 promotes medulloblastoma growth and affects cell cycle progression, microtubule dynamics, and DNA damage response (Abdelfattah et al., 2018). Preclinical studies suggest a role for HDAC1 in the epigenetic regulation of sarcoma tumorigenesis (Bennani-Baiti and Idriss, 2011). HDAC2 is highly expressed in sarcomas (Pacheco and Nielsen, 2012). HDAC3 is a target of the HDAC inhibitors approved by the US Food and Drug Administration to treat lymphoid malignancies (Stengel et al., 2019). HDAC4 overexpression has been observed in acute promyelocytic leukemia (Chauchereau et al., 2004), B cell lymphoma (Sandhu et al., 2012), and multiple myeloma (Kikuchi et al., 2015). HDAC6 is protective in tauopathy and suppresses aberrant tau accumulation. Chronic HDAC6 loss results in accelerated tau pathology, cognitive dysfunction, and reduced survival (Trzeciakiewicz et al., 2020). HDAC7 is a bona fide transcriptional repressor vital to B cell development (Azagra et al., 2016). HDAC8 modulates p53 activity and ensures long-term hematopoietic stem cell maintenance and survival under stress (Hua et al., 2017). HDAC5, HDAC7, and HDAC9 play protective roles during cardiac hypertrophy (Mckinsey, 2011). HDAC11 performs a key function in oncogene-induced, non-homeostatic hematopoiesis (11 deficiency disrupt, 2020).

Nevertheless, the mechanisms by which HDACs are inhibited or promoted in glioma development and progression are unclear. No comprehensive bioinformatics analysis has been conducted to identify the putative role of HDACs in glioma. In the present study, we mined numerous large databases to analyze HDAC expression, mutation, and function and immune infiltration and determine their potential oncogenic and prognostic value in glioma.

Epigenetic modifications are heritable, reversible genetic changes that do not involve DNA mutation. HDACs are epigenetic regulators that remove the acetyl moieties from acetylated lysine residues of post-translationally modified histone proteins that form the core of the chromatin network. This mechanism compresses chromatin, compromises DNA accessibility, and represses transcription and gene silencing (Sengupta and Seto, 2004). However, distinct roles of each HDAC family member in glioma remain to be clarified. In the present study, we analyzed HDAC expression, mutation, prognostic value, functional enrichment, and immune cell infiltration in patients with glioma. To this end, we consulted various public databases including Oncomine, Human Protein Atlas, GEPIA, cBioPortal, CCLE, GeneMANIA, String, and CGGA. GO enrichment and KEGG pathway were analyzed with OmicShare Tools.

Previous studies have evaluated the roles of certain HDAC family members in glioma (Lodrini et al., 2013; Wang et al., 2017; Lo Cascio et al., 2019; Song et al., 2020). A previous study has indicated that the NFAT2-HDAC1 pathway may maintain the malignant phenotype and promote mesenchymal transition in glioma stem-like cells (GSCs). Hence, GSCs are potential molecular targets for GBM therapy (Song et al., 2020). HDAC1 ablation significantly decreased stemness and proliferation in patient-derived GSCs in a p53-dependent manner. By contrast, it had only minimal impact on normal human neural stem cells and astrocytes (Lo Cascio et al., 2019). HDAC1 knockdown inhibited the epithelial–mesenchymal transition transcription factors TWIST1 and SNAIL, downregulated the mesenchymal marker matrix metalloprotein-9, and upregulated the epithelial marker E-cadherin 34. Hence, HDAC1 may contribute to epithelial–mesenchymal transition in glioma cells (Wang et al., 2017). Our results showed that HDAC1 was highly expressed in glioma tumor tissue and significantly correlated with patient survival and immune infiltrate density. Moreover, HDAC1 had the highest mRNA expression level of all HDACs in 12 glioma cell lines. A Cox proportional hazard model of HDACs and clinical factors disclosed that HDAC1 expression was significantly correlated with clinical outcome in LGG. HDAC1 upregulation was significantly correlated with poor OS and DFS in patients with glioma patients. The foregoing associations merit further investigation. We plan to knock out or use the expression of protease inhibitor HDAC1 in glioma cell lines to observe tumor cell invasion and migration in functional experiments. Besides use tumor-bearing nude mice to further verify whether its HDAC1 can inhibit tumor growth.

Recent observations demonstrated that class I HDACs (HDAC1, HDAC2, and HDAC3) modulate DNA damage signaling (Miller et al., 2010; Thurn et al., 2013), maintain genomic stability, and prevent tumorigenesis in vivo (Dovey et al., 2013; Heideman et al., 2013; Santoro et al., 2013). The mammalian class I deacetylase HDAC2 has been extensively studied. HDAC2 downregulation markedly inhibits tumor growth. Thus, HDAC2 may be an oncogene in tumorigenesis (Zhu et al., 2004). HDAC2 protein overexpression was detected in human gastric, prostate, and breast cancers (Nakagawa et al., 2007; Jinwon et al., 2014). HDAC2 represses gene expression by deacetylating H4K16ac (Ma et al., 2013), determines transcription repression, and participates in the nucleosome remodeling deacetylase complex. In vitro and in vivo experiments demonstrated that silencing TRPS1 inhibited tumor growth, whereas HDAC2 overexpression promoted it (Wang et al., 2018). The results of the present study corroborated these findings. HDAC2 expression was higher in glioma tumor than in normal tissues. HDAC2 expression was also significantly correlated with glioma grade and positively correlated with purity infiltration. However, we found that unlike mRNA expression, immune combination expression was elevated in both glioma and normal tissues possibly because of gene regulation during transcription and translation.

During thymocyte development, HDAC3 is required for positive selection and CD4 lineage development (Stengel et al., 2015). HDAC3 represses CD8 lineage genes to keep double-positive thymocytes in a bipotential state (Philips et al., 2019). Here, HDAC3 expression in LGG was positively correlated with the infiltration of B cells, neutrophils, and dendritic cells. HDAC3 mRNA expression was significantly associated with clinical outcome in GBM. Hence, immunity-related cells may be implicated in glioma pathogenesis.

HDAC4 is a key member of class IIa HDACs and performs a wide variety of functions (Seto and Yoshida, 2014). In various cells, HDAC4 is post-transcriptionally regulated by several microRNAs such as miR-1, miR-29, miR-140, miR-155, miR-200a, miR-206, and miR-365 (Wang et al., 2014). Some studies have shown that the molecular mechanism of glioma formation involved HDAC4-mediated SP1 and KLF5 deacetylation in selective MKK7 transcription and oncogenic JNK/c-Jun cascade activation. These findings were consistent with the findings of our research. The results of the immune combination showed that HDAC4 is highly expressed in glioma tumor tissues. KEGG enrichment analysis has revealed that in cancers, HDAC and its neighboring 50 genes are enriched in microRNAs. HDAC4 upregulation was significantly correlated with poor OS and DFS in all patients with glioma. A Cox proportional hazard model of HDAC and its clinical factors disclosed that HDAC4 expression was significantly correlated with clinical GBM outcome.

HDAC5 is a class II HDAC. Aberrant HDAC5 expression has been observed in multiple cancer types. HDAC5 participates in cell proliferation and invasion, immune response, and maintenance of stemness (Yang et al., 2021). In the present study, HDAC5 expression was low in tumor cell lines and in glioma and normal tissues. However, HDAC5 upregulation was significantly correlated with tumor stage, poor OS, and DFS in all patients with glioma.

HDAC6 and 10 are class IIb HDACs based on their subcellular localizations and expression patterns (Haberland et al., 2009). Previous research has shown that HDAC6-selective inhibitors delayed tumor initiation and progression in vivo without causing any significant adverse effects (Auzmendi-Iriarte et al., 2019). HDAC6 inhibition induces an in vivo delay in tumor growth and downregulates PD‐L1 expression. Several key immunological checkpoint modulators are regulated by HDAC6 (Lienlaf et al., 2016). HDAC10 expression is positively associated with PD-L1 expression and may predict the outcome of patients with non-small cell lung carcinoma. HDAC10 inhibition combined with doxorubicin administration kills neuroblastoma but not non-malignant cells by impeding drug efflux and enhancing DNA damage. This novel mechanism targets chemotherapy resistance (Johannes et al., 2018). In this study, HDAC6 expression was higher in glioma tumor than in normal tissues. Low HDAC6 mRNA expression was significantly correlated with OS and DFS in patients with glioma. HDAC10 expression is moderate in LGG and elevated in high-grade gliomas. However, low HDAC10 mRNA expression was not significantly correlated with DFS in patients with glioma.

HDAC7 sustains human mammary epithelial cell proliferation and favors the establishment of stem‐like cell populations by maintaining a proficient microenvironment. High HDAC8 expression levels in human GBM tissues and GBM-R cell lines were correlated with O-methylguanine-DNA methyltransferase levels (Cutano et al., 2019). An inhibitor of HDAC8 combined with temozolomide administration induced WT-p53-mediated apoptosis via WT-p53-mediated O-methylguanine-DNA methyltransferase inhibition in GBM-R cell lines (Tsai et al., 2021). Elevated HDAC9 expression is associated with poor prognosis and promotes malignancy (Huang et al., 2018; Linares et al., 2019; Li et al., 2020). HDAC9 deficiency promoted tumor progression by decreasing CD8⁺ dendritic cell (DC) infiltration in the tumor microenvironment. HDAC9 expression was significantly positively correlated with CD8⁺ cell counts in human lung cancer stroma samples (Ning et al., 2020). In the present study, immune cell infiltration was associated with HDACs in glioma. HDAC9 expression in LGG was positively correlated with CD8⁺ cell infiltration. HDAC7 expression in LGG was positively correlated with the infiltration of CD4⁺ T cells, macrophages, and dendritic cells. HDAC8 expression in LGG and GBM was positively correlated with purity infiltration. High HDAC7 mRNA expression was significantly correlated with OS and DFS in patients with glioma. However, HDAC8 mRNA expression levels were not meaningful for prognostic survival analyses of patients with glioma. HDAC9 mRNA expression levels were not associated with OS in patients with glioma.

HDAC11 is the only member of the class IV HDACs (Seto and Yoshida, 2014). There is limited information regarding its functions. Using a novel class of highly selective HDAC inhibitors and genetically deficient mouse models, the foregoing study revealed that HDAC11 was necessary for oncogenic JAK2-driven myeloproliferative neoplasm cell and tissue proliferation and survival (Yue et al., 2020). Other data demonstrated a significant role of HDAC11 in mitotic cell cycle progression and survival of MYCN-amplified neuroblastoma cells and suggested that HDAC11 is potentially a valuable drug target (Thole et al., 2017). The present study arrived at the same conclusions. KEGG enrichment analysis disclosed that enrichment of HDAC and its neighboring 50 genes was associated with the cell cycle. HDAC11 expression was lower in glioma tumor than in normal tissues. HDAC11 downregulation was significantly correlated with tumor stage, poor OS, and DFS in patients with glioma.

There are several limitations to our study. Firstly, we only analyzed data retrieved from online databases, we must validate our findings using large cohorts. Secondly, our HDAC IHC and immunofluorescence data were incomplete. Third, we did not use our cell lines or tissues to further verify the results. Although HDAC family expression data are similar for glioma cell lines and pancancer analyses. Perhaps, Basic experiments must be conducted to elucidate the mechanisms of HDAC family members and determine their impact on epigenetics. At present, a large number of HDAC inhibitors have been used in other tumors, and we can use HDAC inhibitors to act on cells or animals to further observe their inhibition of tumors. The conventional histone acetylases show lack of blood-brain barrier permeability in the treatment of brain diseases, but through the development of new small molecule probes and carriers, they can penetrate the blood-brain barrier well. we may use nano-encapsulation to further break through the blood-brain barrier (Dang, 2014). In the actual human clinical safety and effectiveness test, we can observe the immunohistochemical expression of HDAC in clinical work and combine with traditional glioma gene diagnosis to analyze and judge the prognosis of glioma diseases.

Most of these histone Acetylases are usually adjunctive to other chemoradiotherapy drugs such as temozolomide/bevacizumab. However, most of the current clinical trials are phase I/II clinical trials, and further phase III trials are needed to demonstrate the role of histone acetylases in gliomas. In a phase two Study of Concurrent Radiation Therapy, Temozolomide, and the Histone Deacetylase Inhibitor Valproic Acid for Patients With Glioblastoma, histone inhibitors may improve outcomes for patients with glioma compared with historical data (Krave, 2015). In the present study, we systematically analyzed HDAC expression and prognostic value in glioma and revealed the heterogeneity and complexity of the molecular mechanisms associated with these cancers. We also investigated the mRNA expression patterns, prognostic values, genetic alterations, GO enrichment, and PPI networks of HDACs in patients with glioma. According to the immune prognosis analysis, the age and HDAC1 mRNA expression of patients with LGG were related to disease prognosis. Patient age, CD4⁺ T cell density, and HDAC3 and HDAC4 mRNA expression were associated with high-grade glioma prognosis. HDAC1 and HDAC2 were relatively upregulated whereas HDAC11 was downregulated in LGG and GBM tumor tissues. High HDCA1 and low HDAC11 mRNA expression levels were significantly related to glioma OS and DFS. HDAC2 mRNA upregulation had prognostic value in OS. The foregoing results indicate that HDAC1/2 are potential prognostic biomarkers for patients with glioma. The present study may facilitate the discovery of novel prognostic biomarkers for glioma among other HDAC family members.