Histone deacetylases 10 as a prognostic biomarker correlates with tumor microenvironment and therapy response in colorectal cancer

Abstract

BACKGROUND

The histone deacetylases 10 (HDAC10) is a HDAC family member, yet its importance in the context of colorectal cancer (CRC) development remains incompletely understood. The present study was thus developed to explore the mechanistic importance of HDAC10 as a regulator of CRC.

AIM

To investigate the impact of HDAC10 on tumor growth and its regulation in tumor microenvironment (TME) in CRC, we conducted this study.

METHODS

The study evaluated HDAC10 expression using immunohistochemistry analyses and assessed its prognostic value in CRC patients. HDAC10 depletion CRC cell lines were generated, and its biological functions were assessed through cell counting kit-8, wound healing, and colony formation assays. Furthermore, gene set variation analysis (GSVA) was employed to explore the potential molecular mechanisms of HDAC10 in CRC. The impact of HDAC10 on TME was subsequently assessed. Finally, the study investigated the influence of HDAC10 on the response to immunotherapy and chemotherapeutic drugs in CRC.

RESULTS

HDAC10 expression was significantly elevated in CRC and correlated with poor prognosis in patients. Knockdown of HDAC10 reduced colon cancer cell proliferation and migration capabilities. GSVA revealed a strong association between high HDAC10 expression and immune suppression. Additionally, high HDAC10 levels were correlated with a non-inflamed TME. Finally, patients with high HDAC10 expression showed reduced sensitivity to immunotherapy.

CONCLUSION

This study revealed the significance of HDAC10 in TME, therapy efficacy, and clinical prognosis in CRC, offering novel insights for therapeutic advancements in CRC.

Key Words: Histone deacetylases 10; Colorectal cancer; Tumor microenvironment; Immunotherapy; Prognosis

Core Tip: Histone deacetylases 10 (HDAC10), a HDAC family member, was overexpressed in colorectal cancer (CRC) and associated with poor prognosis. HDAC10 depletion inhibited CRC cell proliferation and migration, suggesting its oncogenic role. Gene set variation analysis linked elevated HDAC10 expression to immunosuppressive pathways and a non-inflamed tumor microenvironment (TME), characterized by reduced immune cell infiltration. CRC patients with high HDAC10 levels showed reduced sensitivity to immunotherapy, indicating therapeutic resistance. Our findings highlight HDAC10’s critical role in shaping the TME, treatment response, and clinical outcomes in CRC, supporting its potential as a therapeutic target to improve CRC management.

INTRODUCTION

Colorectal cancer (CRC) ranks as the third most prevalent cancer globally and stands as the second primary cause of cancer-related mortality worldwide, thereby posing a significant global health challenge and underscoring the pressing demand for innovative therapeutic approaches[1]. Although novel treatment modalities including targeted therapy and immunotherapy have been developed, the average five-year survival rate for individuals with advanced CRC still less than 15%[2,3]. Furthermore, patients with oligometastatic disease following tumor resection and systemic therapy have a five-year survival rate of only 40%[4-7]. Given the limitations of current therapies, there is an urgent need to identify more effective biomarkers for prognostic prediction and personalized treatment[8].

Post-translational modifications (PTMs) are indispensable for maintaining protein function by regulating the activity, stability, and localization of proteins[9]. Among major PTMs, deacetylation and acetylation are significant factors in signaling and cellular metabolism[10]. Histone acetylation and deacetylation are typically regulated by histone acetyltransferases and histone deacetylases (HDACs), respectively. Dysregulation of these processes is implicated in various diseases, including cancer[11]. Notably, histone acetylation and deacetylation have been found to modulate the host immune response in cancer and regulate anti-tumor immune responses, thereby affecting the efficacy of immunotherapy[12].

The HDAC family, a histone deacetylase subfamily, includes HDAC1-11[13]. It plays an important regulatory role in immune cells and also participates in immunomodulation in cancer cells[14]. Previous studies have demonstrated that HDAC inhibitors increase PD-L1 expression in tumor cells, thereby enhancing the immunotherapy efficacy[15]. HDAC10, a member of the HDAC family, has been found to promote the inhibition of immune response by immune regulatory cells[16]. Recent research has reported that HDAC10 could modulate critical biological processes in ccRCC (clear cell renal cell carcinoma), including proliferation, migration, and apoptosis[17]. In addition, HDAC10 inhibition repressed melanoma cell growth and BRAF inhibitor resistance via upregulating SPARC expression[18]. Furthermore, knockdown of HDAC10 inhibited the proliferation of non-small cell lung carcinoma (NSCLC) cells and promoted their ferroptosis by regulating the SP1/POLE2 axis[19]. However, the impact of HDAC10 on cancer cells and tumor immunity in CRC has not been reported.

In this study, we found that HDAC10 promoted cancer growth and metastasis, inhibited the immune response, and reduced immunotherapy efficacy.

MATERIALS AND METHODS

Patients and tissue samples

Thirty CRC tissues and paired normal tissues were obtained from Zhongnan Hospital of Wuhan University between January 2024 and April 2024 with patient informed consent. The study was reviewed and approved by the Medical Ethics Committee of the Zhongnan Hospital of Wuhan University Institutional Review Board [(Approval No. 2025001K)]. A CRC tissue microarray, including 90 CRC samples, was obtained from Shanghai Outdo Biotech Company. All the patients were classified according to the 8th edition of TNM staging system released by the Union for International Cancer Control and the American Joint Committee on Cancer. Written informed consent was obtained from all participants. Their clinicopathological characteristics are presented in Supplementary Table 1.

Immunohistochemistry

In brief, CRC and normal tissues from a tissue microarray were fixed in 40 g/L neutral-buffered formalin (Sigma-Aldrich, United States) and embedded in paraffin. Tissue sections were sliced from paraffin blocks into 3-μm-thick slices. Tris-EDTA buffer (1.0 mmol/L; pH 9.0) was used to perform heat mediated antigen retrieval. Primary antibodies used included HDAC10 (1:200, ELK Biotechnology, Wuhan, China, ES2519), CD8 (1:200, Proteintech, Wuhan, China, 66868-1-Ig), CXCL9 (1:200, Proteintech, Wuhan, China, 22355-1-AP), and CXCL10 (1:200, Proteintech, Wuhan, China, 10937-1-AP). Images were processed with Image J software, and relative expression was calculated.

Cell culture and transfection

Two human CRC cell lines, SW480 and HCT116, were purchased from the China Center for Type Culture Collection (Wuhan, China). Fetal bovine serum (FBS) (HyClone, United States) is heat-inactivated (56 °C water bath for 30 minutes) before use. Because it can remove some growth factors and complement and reduce variation between batches, thus ensuring the stability and reliability of experimental results. The cells were cultured in DMEM (GIBCO, United States) supplemented with 40 g/L FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin (Genom, China), at 37 °C with 50 mL/L CO₂. siHDAC10 and FAM-labeled siNC was acquired from Guangzhou RiboBio (Guangzhou, China) and subsequently transfected into CRC cell lines by using Lipofectamine 3000 (Invitrogen, United States). The sequences of siNC and siHDAC10 are shown: SiNC: ACGUGACACGUUCGGAGAA; siHDAC10-1: GACAGTTCGACGCCATCTA; siHDAC10-2: GTGGTTTCCTGAGCTGCAT. After 24 hours, the transfection efficiency of small interfering RNA (siRNA) was assessed with a fluorescence microscope (Olympus U-RFL-T, manufactured in Japan). Furthermore, the knockdown efficacy of siHDAC10 was confirmed through quantitative real-time PCR (qRT-PCR) and western blotting analyses.

RNA extraction and qRT-PCR analysis

Total RNA in tissues and cells was extracted using Trizol reagent (Invitrogen, United States), followed by reverse transcription with the TOYOBO ReverTra Ace kit (TOYOBO, Japan). qRT-PCR was performed using UltraSYBR mixture (Cwbio, China) on a Roche LightCycler 96 PCR system. mRNA expression levels were quantified via qRT-PCR using the Roche LightCycler 96 PCR system, and GAPDH was utilized as the internal control gene. The primers were synthesized by TSINGKE Biological Technology (Wuhan, China), and their specific sequences were presented as follows: GAPDH Forward: 5′-GGAGCGAGATCCCTCCAAAAT-3′ and Reverse: 5′-GGCTGTTGTCATACTTCTCATGG-3′; HDAC10 Forward: 5′-CCTAGAGTCCATCCAGAGT-3′ and Reverse: 5′-GCTGCTATACCACTGTTCA-3′ and p53 Forward: 5′- GAGGATTCACAGTCGGATA-3′ and Reverse: 5′-ATCATCTGGAGGAAGAAGTT-3′. The quantification of mRNA expression levels was conducted utilizing the comparative CT method (2^-ΔΔCT), and all experimental procedures were conducted in triplicate.

Protein extraction and western blotting

The total protein from CRC cell lines was extracted utilizing RIPA lysis buffer (Boyotime, China) according to manufacturer's instructions. western blotting was performed with the specific antibody, HDAC10 (1:1000, ELK Biotechnology, Wuhan, China, ES2519), GAPDH (1:1000, Proteintech, Wuhan, China, 60004-1-Ig) and p53 (1:1000, Cell Signaling Technology, American, 2524T).

Cell counting kit-8

Cell viability assessment was carried out using the Cell counting kit-8 (CCK-8) kit (Dojindo Molecular Technologies, Japan). Specifically, 5000 cells, which had been transfected with either siHDAC10 or siNC, were seeded into each well of 96-well plates. Each experimental condition was replicated across five wells to ensure data reliability. Subsequently, the CCK-8 reagent was added to the wells along with the cell culture medium, followed by a 2-hour incubation period before measurement. Absorbance readings at 450 nm were obtained using a microplate reader (ELX-800; BioTek, United States), and the growth curve was plotted based on the optical density values. The experiments were performed in triplicate.

Colony formation assay

After 24 hours of siRNA transfection, 5000 cells were plated into a six-well culture plate and incubated at 37 °C in an atmosphere containing 50 mL/L CO₂. Four days post-transfection, a second round of transfection was performed on these cells. Colonies became visible after 10 days of incubation. Subsequently, the colonies were fixed using 40 g/L paraformaldehyde and stained with crystal violet. The colonies comprising a minimum of 50 cells were then counted and quantified. The experiments were performed in triplicate. Images were processed with Image J software, and relative expression was calculated.

Wound-healing assay

HCT116 or SW480 cells were seeded, and scratch wounds were made when cell confluence reached approximately 80% at approximately 48 hours post-transfection. Subsequently, the cells were incubated in serum-free medium for 24 hours. Following this, three random visual fields were selected from each scratch wound and observed under a microscope. The experiments were performed in triplicate. Images were processed with Image J software, and relative expression was calculated.

Data acquisition and processing

The TIMER database (http://timer.comp-genomics.org/timer/) was utilized to analyze HDAC10 expression levels in pan-cancer data. The gene expression matrix and clinical information of CRC patients were obtained from The Cancer Genome Atlas Program (TCGA, https://www.cancer.gov/ccg/research/genome-sequencing/tcga), as well as from GSE39582 cohort in the Gene-Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). Gene Expression Profiling Interactive Analysis 2 (http://gepia2.cancer-pku.cn/) was exploited to investigate the expression and prognostic values of HDAC10 in CRC. Batch effects from non-biological technical biases were corrected using the “ComBat” algorithm of “sva” R package. All RNA-seq data were standardized as log2 (TPM + 1) transformation using appropriate R packages before analysis.

Development and assessment of a nomogram

Univariate Cox (Uni-Cox) and multivariate Cox (Multi-Cox) regression analyses were employed to assess the prognostic value of HDAC10 in CRC. The “rms” R package was utilized to construct nomogram model and calibration curve. Receiver operating characteristic (ROC) curve analysis was carried out to determine overall survival at the 1-, 3-, and 5-year time points using “timeROC” R package. The area under the curve values were calculated to evaluate the sensitivity and specificity of HDAC10 as a prognostic biomarker in CRC. Additionally, decision curve analysis was performed using the "rmda" R package.

Differentially expressed genes screening

The TCGA-COAD patients were divided into high and low HDAC10 expression groups based on the median HDAC10 expression level as the cutoff. The "limma" R package was utilized to compute the P value and log2 fold changes (FC) for whole genome genes when comparing patients with high HDAC10 expression to those with low HDAC10 expression. The differentially expressed genes (DEGs) were identified based on the criteria of an adjusted P value < 0.05 and |log2 FC| > 0.5.

Gene set variation analysis

The "GSVA" R package was employed to conduct gene set variation analysis (GSVA) analysis, aiming to elucidate the differences in biological processes between the high-HDAC10 and low-HDAC10 groups.

Analysis of correlation between HDAC10 and tumor microenvironment infiltration

To calculate the ESTIMATE score, immune score, stromal score, and tumor purity for CRC using the "ESTIMATE" R package. In addition, a Pearson correlation analysis was conducted to examine the association between HDAC10 expression levels and tumor purity. Moreover, the "linkET" R package was employed to investigate the correlation between HDAC10 expression and the cancer-immunity cycles. The single-sample gene set enrichment analysis (ssGSEA) scores for 28 different immune cell types were computed using the ssGSEA algorithm. The infiltration levels of tumor-infiltrating immune cells were calculated using seven algorithms available in the TIMER database[20,21].

Immunotherapy and drug sensitivity analysis

To evaluate the predictive ability of HDAC10 in response to immunotherapy, eight immunotherapy cohorts were downloaded from the GEO and Tumor Immune Dysfunction and Exclusion (http://tide.dfci.harvard.edu/Login/). In addition, the IMvigor210 cohort were obtained from (http://research-pub.Gene.com/imvigor210corebiologies/). The "oncoPredict" R package was utilized to calculate the semi-inhibitory concentration (IC₅₀) values of chemotherapeutic drugs.

Tumor somatic mutation analysis

Initially, the "maftools" R package was employed to collect and examine somatic mutation data obtained from CRC patients. Subsequently, a visual analysis was conducted on the ten genes exhibiting the highest tumor mutation frequencies in each HDAC10 subgroup. Finally, the tumor mutation burden (TMB) was computed for every sample to investigate the variations in TMB levels across the different HDAC10 subgroups.

Statistical analysis

All statistical analyses were conducted utilizing R software (version 4.1.3) along with GraphPad Prism 8.0 (GraphPad, San Diego, CA, United States). For pairwise comparisons, the Wilcoxon test was applied, whereas the Kruskal-Wallis test was used for multiple group comparisons. Kaplan-Meier survival curves were constructed and subsequently analyzed through the "survival" R package and the logrank test, respectively. Measurement data are presented as the mean ± SD, and P value < 0.05 was considered to indicate a significant difference.

RESULTS

Higher HDAC10 expression in CRC and was associated with poor prognosis

In order to identify which HDACs was significantly upregulated in CRC, we examined the expression of HDACs in TCGA-COAD cohort. The results showed that HDAC10 expression was significantly upregulated in CRC than paired paracancerous tissues (Supplementary Figure 1). Therefore, we chose HDAC10 as our research core. Firstly, in pan-cancer analysis, HDAC10 expression was significantly upregulated in multiple cancers, including colon, lung, esophagus (Figure 1A). Further investigation of HDAC10 expression demonstrated a comparable elevation in CRC tumor tissues when contrasted with normal colon tissues (Figure 1B; Supplementary Figure 2A-C). To validate the potential role of HDAC10 in CRC, we conducted qRT-PCR analysis on thirty CRC patient samples, showing consistent results (Figure 1C). Additionally, immunohistochemistry (IHC) analysis of CRC tissue microarray demonstrated that HDAC10 expression was significantly higher in CRC than paired paracancerous tissue (Figure 1D and E). Furthermore, HDAC10 was found to predict patients prognosis and correlated with poor prognosis (Figure 1F and G), which was validated in external datasets (Supplementary Figure 2D-I). In summary, HDAC10 is a potential prognostic marker in CRC.

Figure 1 Higher histone deacetylases 10 expression in colorectal cancer and was associated with poor prognosis. A: Higher histone deacetylases 10 (HDAC10) expression in normal and tumor tissues in pan-cancer data from the TIMER website; B: HDAC10 expression analysis in paired colorectal cancer (CRC) samples from TCGA-COAD showed high expression in tumor; C: Quantitative real-time polymerase chain reaction (qRT-PCR) analysis from thirty patients tissues indicated HDAC10 upregulated in CRC; D: Representative cases in normal tissues and CRC showed high HDAC10 expression in CRC. Scale bar, 100 μm and 20 μm; E: Quantification of immunohistochemistry (IHC) positive areas in normal and CRC indicated high levels of HDAC10 in CRC; F: Receiver operating characteristic curve based on HDAC10 for predicting 1-, 3- , 5-year overall survival; G: Kaplan-Meier survival analysis of HDAC10 expression in CRC tissue microarray and TCGA-COAD datasets showed patients with expressing higher levels of HDAC10 had poor prognosis. ^aP < 0.001, ^bP < 0.01, ^cP < 0.05 vs the NC group. The statistical power for comparing the rates between NC group and CRC group in qRT-PCR and IHC experiments using PASS software, and the results were 71% and 73%, respectively. CRC: Colorectal cancer; OS: Overall survival; TCGA: The Cancer Genome Atlas; HDAC10: Histone deacetylases 10; AUC: Area under the curve.

HDAC10 was associated with clinicopathological characteristics and nomogram model construction

In the correlation analysis between HDAC10 and clinicopathological factors in CRC tissue microarray, we observed that elevated HDAC10 was correlated with advanced T/N/M stage, pathological stage and higher grade (Figure 2A-H). These results were validated in TCGA datasets (Supplementary Figure 3A-H). Considering HDAC10 as a potential prognostic marker for CRC patients, we conducted Uni-Cox and Multi-Cox regression analyses, revealing HDAC10 as an independent prognostic factor for CRC patients (Figure 2I-J). Further analysis revealed that HDAC10 served as a reliable prognostic indicator for patients at 1-, 3-, and 5-years through the construction of a nomogram model (Figure 2K). Calibration curves also supported this conclusion (Figure 2L). Moreover, the prognosis prediction of model was accurate (Figure 2M) and provided significant clinical benefits for patients (Figure 2N). These results were also verified in TCGA datasets (Supplementary Figure 3I-K).

Figure 2 Histone deacetylases 10 was associated with clinicopathological characteristics and nomogram model construction. A: Differential analysis for histone deacetylases 10 (HDAC10) expression in patients with different age; B: Gender; C: High HDAC10 expression in high grade; D: Status; E: High levels of HDAC10 in advanced T stage; F: High HDAC10 expression in advanced N stage; G: High HDAC10 expression in advanced M stage; H: High levels of HDAC10 in advanced pathologic stage; I and J: Univariate Cox and multivariate Cox regression analysis for clinical factors and HDAC10 expression with overall survival (OS). Hazard ratio > 1 represented risk factors for survival and hazard ratio < 1 represented protective factors; K: Nomogram for predicting the 1-, 3-, and 5-year OS from CRC tissue microarray; L: Calibration curves for predicting 1-, 3-, and 5-year OS; M: Receiver operating characteristic curve for predicting 1-, 3-, and 5-year OS; N: Decision curve analysis based on nomogram better determine survival than only HDAC10. OS: Overall survival; HDAC10: Histone deacetylases 10; AUC: Area under the curve.

HDAC10 promote tumor proliferation and migration

To delve deeper into the biological implications of HDAC10 in CRC, we employed siRNAs (siHDAC10-1 and siHDAC10-2) to suppress HDAC10 expression in CRC cell lines. We preliminarily evaluated the siRNA transfection efficiency in CRC cells through fluorescence image 24 hours after transfection (Supplementary Figure 4A), qRT-PCR and western blotting were performed to verify the silencing efficiency of HDAC10 (Figure 3A). The CCK-8 assay showed that HDAC10 knockdown significantly inhibited cancer cell proliferation (Figure 3B). Similarly, the colony formation assay demonstrated that HDAC10 knockdown inhibited CRC cell proliferation (Figure 3C). IHC staining also showed that Ki67 expression was significantly suppressed in the low HDAC10 group (Supplementary Figure 4B). It is well-known that p53 served as a crucial tumor suppressor mechanism, which inhibited cell proliferation in human malignancies by transcriptionally activating target genes[22]. Therefore, we further explore the regulation of HDAC10 on p53. The results showed that HDAC10 depletion increased p53 expression at transcriptional and translational levels (Supplementary Figure 4C). Moreover, the wound healing assay demonstrated that HDAC10 knockdown markedly attenuated the migration of CRC cells when compared with control cells (Figure 3D). To summarize, HDAC10 knockdown impairs the proliferation and migration capabilities of CRC cells.

Figure 3 Histone deacetylases 10 promote tumor proliferation and migration. A: Knockdown of histone deacetylases 10 (HDAC10) expression in SW480 and HCT116 cells at mRNA and protein levels; B: Cell counting kit-8 assay showed HDAC10 depletion inhibited cell proliferation; C: Colony formation analysis indicated HDAC10 knockdown suppressed colony formation; D: Wound-healing assay showed HDAC10 depletion inhibited cell migration. Compared with the siControl group, ^aP < 0.001. Data were represented as mean ± SD of biological triplicates. HDAC10: Histone deacetylases 10.

Differential analysis and functional enrichment analysis

A differential expression analysis was conducted to compare samples exhibiting high HDAC10 expression with those showing low HDAC10 expression, resulting in the identification of 1828 DEGs (Figure 4A; Supplementary Table 2). Among these DEGs, the top five significantly downregulated and upregulated genes displayed evident differences between the two groups (Figure 4B). Notably, recent research revealed that PABPC1 L facilitates immune evasion in renal cell carcinoma[23], while CLDN18 and CTSE promote antitumor immunity[24,25]. Additionally, principal component analysis confirmed the separation of the two groups based on HDAC10 expression (Figure 4C). To further explore the potential functional mechanisms of HDAC10 in CRC, GSVA was performed, showing that the low HDAC10 group was enriched in immune activation pathways, such as IL2-STAT5 and IFN-γ signaling[26,27]. In contrast, high HDAC10 expression was closely related to immunosuppressive pathways, including PI3K-AKT-mTOR[28,29], and Wnt/ß-catenin signaling[30] (Figure 4D; Supplementary Table 3).

Figure 4 Differential analysis and functional enrichment analysis. A: Volcano plot from the top five upregulated and downregulated differentially expressed genes (DEGs) between high and low-histone deacetylases 10 group; B: Heat map showed the top five genes in both upregulated and downregulated DEGs; C: Different expression levels of DEGs between normal colon and colorectal cancer; D: Gene set variation analysis showed the activation status of biological pathways in two groups. HDAC10: Histone deacetylases 10.

Correlation between HDAC10 and TME

While identifying that HDAC10 enhanced tumor malignancy and predicted patients’ prognosis, a comprehensive analysis of tumor microenvironment (TME) was performed. Firstly, high HDAC10 expression closely correlated with low infiltration levels of immunomodulatory factors, including MHC, immunostimulator, chemokine, receptor (Figure 5A). Then, the ESTIMATE, immune, and stromal scores were lower in high HDAC10 group than low group (Figure 5B), and HDAC10 was positively correlated with tumor purity (Figure 5C). Subsequently, the infiltration levels of immune cells including CD8+ T cell, macrophage, and NK cell were higher in the low HDAC10 group than in the high group (Figure 5D). These results were validated in GEO datasets (Supplementary Figure 5). Furthermore, as determined by seven distinct algorithms, the expression of HDAC10 exhibited a negative correlation with the infiltration levels of CD8+ T cells, macrophages, and NK cells (Figure 5E). Considering the crucial function of CD8+ T cells in anti-tumor immune responses and their observed relationship with HDAC10 expression, IHC analysis was subsequently performed on the collected colon cancer tissue. The results demonstrated a notable upregulation of CD8a, a hallmark marker of CD8+ T cells, in tissues exhibiting low HDAC10 expression. Furthermore, CXCL9, CXCL10 has been demonstrated to enhance the accumulation of effector T cells at the tumor site and suppress tumor growth[31]. Likewise, in our tissue group, we noted an increased expression of CD8a, CXCL9, and CXCL10 in tissues that exhibited low HDAC10 expression (Figure 5F). Finally, HDAC10 expression was negatively related to critical steps in CI cycle, such as CD8+T cell and macrophage recruitment, as well as recognition and killing cancer cells (Figure 5G). In summary, low HDAC10 expression is associated with high immune infiltration.

Figure 5 Correlation between histone deacetylases 10 and tumor microenvironment in The Cancer Genome Atlas database. A: Low expression of 4 immunomodulators (MHC, immunostimulators, chemokines and receptors) in high and low-histone deacetylases 10 (HDAC10) groups; B: Low stromal, immune, and ESTIMATE scores in high HDAC10 groups; C: HDAC10 expression positively correlated with tumor purity; D: Low infiltration level of common immune cells in high HDAC10 group; E: Correlation between HDAC10 expression and infiltration levels of tumor-infiltrating immune cells calculated by seven independent algorithms; F: Immunohistochemistry analysis showed that low expression of CD8a, CXCL9 and CXCL10 in high HDAC10 group. Scale bar, 100 μm; G: HDAC10 expression was negatively associated with critical steps in cancer-immunity cycle. ^aP < 0.001, ^bP < 0.01, ^cP < 0.05. NS: No significance; HDAC10: Histone deacetylases 10.

Effect on immune checkpoints and TMB

To further explore the potential correlation between HDAC10 and immunotherapy in CRC, we initially analyzed the expression of common immune checkpoints. This analysis revealed that CD274 (PD-L1), CD28, CD48, and IDO1 were lower in the high HDAC10 group compared to the low HDAC10 group (Figure 6A). Additionally, correlation analysis showed that PD-L1 expression was negatively associated with HDAC10 in CRC tissue microarray (Figure 6B). Next, we investigated the significant impact of TMB on immunotherapy and found that the TMB score was higher in the low HDAC10 group compared to the high HDAC10 group (Figure 6C). This result suggests that low HDAC10 levels expose a higher number of potentially immunogenic neoantigens, thereby enhancing immunotherapy efficacy[32]. Finally, we analyzed the effect of TMB on immunotherapy from a genetic perspective and observed that the TP53 mutation rate was significantly higher in the low HDAC10 group compared to the high HDAC10 group (Figure 6D). Previous studies have confirmed that TP53 mutations represent a state of high immunogenicity and increased sensitivity to immunotherapy[33]. Taken together, high HDAC10 expression is significantly associated with low immune checkpoints expression and low TMB.

Figure 6 Effect on immune checkpoints and tumor burden mutation. A: Low expression of immune checkpoints in high histone deacetylases 10 (HDAC10) group; B: HDAC10 expression negatively correlated with PD-L1 in colorectal cancer tissue microarray; C: Low tumor burden mutation score in high HDAC10 group; D: Waterfall plot for somatic mutations in different HDAC10 groups. ^aP < 0.001, ^bP < 0.01, ^cP < 0.05. HDAC10: Histone deacetylases 10.

Immunotherapy and drug sensitivity analysis

Considering the correlation between HDAC10 and immune checkpoints and TMB, we further investigated the association between HDAC10 and immunotherapy. We found that the efficacy of common immunotherapies, including anti-PD-1, anti-PD-L1, and anti-CTLA-4 treatments, was higher in the low HDAC10 group compared to the high HDAC10 group (Figure 7A-H). While identifying that high HDAC10 expression reduced the benefit of immunotherapy, we also observed that the responsiveness to common chemotherapy drugs-including cisplatin, epirubicin, irinotecan, vincristine, and gemcitabine-was poorer in the high HDAC10 group compared to the low HDAC10 group (Figure 7I-P). In summary, low HDAC10 expression is closely related to high immunotherapy and chemotherapy response.

Figure 7 Immunotherapy and drug sensitivity analysis. A-H: Low immunotherapy response in high histone deacetylases 10 (HDAC10) group from Gide 2019, GSE53127, GSE91061, GSE100797, GSE103668, GSE111636, GSE126044, and IMvigor210 cohorts; I-P: Low common chemotherapy drug sensitivity in high HDAC10 group. HDAC10: Histone deacetylases 10.

DISCUSSION

In recent years, HDACs have been reported to be involved in multiple stages of cancer, and their presence is associated with advanced disease and poor patient outcomes. Additionally, research has shown that HDACs can modulate immune activity and response in various ways, thereby influencing the efficacy of immunotherapy[34]. High levels of HDAC10 have been linked to high PD-L1 expression and poor prognosis in NSCLC patients[35]. However, the potential role of HDAC10 in CRC has not been reported, which piqued our interest. In this study, we found that high HDAC10 expression promotes CRC cell proliferation and migration, and is associated with poor prognosis. More importantly, HDAC10 regulates TME and reduces the benefit of immunotherapy.

In recent years, tumor molecular classification has emerged as a critical area of research. This method not only helps in distinguishing between various tumor subtypes but also offers valuable insights into tumor progression and prognosis. Consequently, it enables the development of personalized treatment plans and enhances the ability to predict patient responses to different therapies[36]. In the context of CRC, proficient mismatch repair or microsatellite stability (pMMR/MSS) status serves as a reliable predictor of immunotherapy efficacy. Our previous research has demonstrated that pMMR/MSS CRC exhibited significant immunogenic heterogeneity, suggesting a propensity for resistance to immunotherapy. Additionally, we have identified GBP2 as a prognostic biomarker for pMMR/MSS CRC patients, influencing both prognosis and immune response[37]. Similarly, our current study underscored the potential utility of HDAC10 in the molecular classification of CRC. We observed that HDAC10 was highly expressed in MSS/MSI-L subtype CRC and was correlated with low differentiation and lymphatic metastasis. Moreover, elevated HDAC10 expression was associated with a worse prognosis and resistance to immunotherapy in CRC patients. Therefore, a comprehensive exploration of the molecular and functional mechanisms underlying HDAC10's role in CRC is essential for elucidating its significance in molecular subtyping research.

In analysis of clinical pathological information from patients using tissue microarray, along with our experiments on tumor cell proliferation and migration, we identified that HDAC10 depletion significantly reduced the proliferation and migration abilities of CRC cells. Additionally, original research has reported that HDAC10 inhibitors (HDAC10i) modulated autophagy in aggressive FLT3-ITD positive AML (acute myeloid leukemia) cells[38]. Subsequent study further demonstrated that HDAC10i showed dose-dependent growth inhibition of HeLa cells[39]. Furthermore, HDAC10-specific inhibitors DKFZ-728 and DKFZ-748 were demonstrated to be an antitumor strategy that have value in inhibiting HCT116 cells proliferation[40]. The conclusion was consistent with our findings. It is commonly known that the activation of the p53 tumor suppressor can trigger cell cycle arrest[41]. In this research, we identified that HDAC10 knockdown increased p53 expression at transcriptional and translational levels. Some research has reported that HDAC2 accelerated cell cycle progression by inhibiting p53 and promoting MYC expression, contributing to malignancy[42]. This indicated that HDAC10 may promoted cell cycle through p53 inhibition, leading to cell proliferation.

Enrichment function analysis indicated that HDAC10 is closely associated with the PI3K-AKT-mTOR and Wnt/β-catenin pathways. Previous studies have demonstrated that HDAC10 promote cancer progression via these signaling pathways[43,44]. In addition, recent studies have reported that HDAC3 deteriorated CRC progression via microRNA-296-3p/TGIF1/TGFβ axis and HDAC5 promoted CRC cell migration by negatively regulating miR-148a-3p[45,46]. Our differential expression analysis showed significant downregulation of REG1A, CLDN18, and MMP13 in the HDAC10 high-expression group. REG1A, known as a Wnt pathway inhibitor[47], and Qiu et al[48] has reported that REG1A suppress tumor growth by inhibiting the PI3K-AKT-mTOR pathway. In gastric cancer, the loss of CLDN18 is linked to the activation of the Wnt pathway and resistance to PD-L1 therapy[49]. MMP13 has been found to inhibit the activation of both the Wnt/β-catenin pathway[50] and the PI3K-AKT-mTOR pathway[51]. These findings suggest that the potential biological functions of HDAC10 are worth further investigation. Building on our current study, we aim to explore the biological roles and regulatory mechanisms of HDAC10 in colon cancer more comprehensively.

Chemokines are a group of relatively small-molecular-weight secreted proteins that induce immune cell movement and function by interacting with chemokine receptors[52]. CXCL9 and CXCL10, in particular, play important roles in antitumor immune responses. Our correlation and IHC analysis showed that high HDAC10 expression was associated with low expression of CXCL9 and CXCL10 in CRC. It has been reported that the HDAC inhibitor HPTA increases CXCL9 and CXCL10 expression and recruits CXCR3+CD4+ T cells to enhance the immune response by activating the NF-κB pathway in breast cancer[53]. Additionally, HDAC3 suppresses CXCL10 expression by directly binding and deacetylating the CXCL10 promoter, thereby inhibiting antitumor immunity in hepatocellular carcinoma (HCC)[54]. Therefore, as a member of the HDAC family, we reasonably speculate that HDAC10 may inhibit the immune response by regulating CXCL9 and CXCL10 in CRC.

Some research has indicated that the resistance to immune checkpoint inhibitor (ICI) treatment is presumably due to the immunosuppressive nature of the TME[55]. In this study, we identified that high expression of HDAC10 significantly correlated with low infiltration levels of CD8+T cell. As we all know, increased CD8+ T cell can strengthen the anti-tumor immune response[56]. As reported, HDAC2 recruited STAT1 to the promoter of ACKR3 to activate its transcription, thus inducing M2 macrophage migration and reducing infiltration level of CD8+T cells[57]. Therefore, as part of HDACs family, it is worth exploring potential mechanisms through which HDAC10 modulates CD8+ T cells and TME in CRC, as a critical pathway for future research.

ICIs have emerged as a primary modality in the treatment of various cancers, including melanoma, lung cancer, and many other solid tumors, showcasing significant clinical efficacy[58]. Interestingly, inhibiting HDAC expression in combination with ICIs represents a promising therapeutic strategy[59]. Recent findings indicate that combining HDAC8 expression inhibition with a PD-L1 antibody in treating aggressive HCC yielded remarkable outcomes[60], and HDAC6 inhibitor ACY-1215 enhanced STAT1 acetylation to block PD-L1 for enhancing CRC immunotherapy[61]. These conclusions are consistent with our findings. Downregulated HDAC10 expression was related to high expression of PD-L1/PD-1 and better immunotherapy response in PD-L1/PD-1 therapy cohorts. Meanwhile, our analysis concerning the impact of TMB on immunotherapy indicated that low HDAC10 expression was associated with high TMB. Previous research has demonstrated that high TMB is associated with a higher number of potentially immunogenic neoantigens, thereby enhancing the immune response[32]. Furthermore, we found that low HDAC10 expression was positively associated with a high TP53 mutation rate. TP53 mutation has been reported to represent a state of high immunogenicity, thus contributing to probable sensitivity to PD-1 blockade[33]. It is reported that p53 can regulate adaptive immune responses by modulating the expression of PD-L1[62]. In addition, we confirmed that HDAC10 depletion could increase p53 expression. Therefore, we consider that p53 as a critical breakthrough point for further exploring the enhancement of combined immunotherapy efficacy in CRC by targeting HDAC10.

While the study results generally met expectations, several limitations were identified. Due to the lack of high-quality fresh CRC samples, the TME analysis primarily relied on publicly available databases, which made it impractical to employ techniques like flow cytometry and single-cell sequencing. Moreover, the construction of cellular or animal models to investigate the effects of HDAC10 on the TME has yet to be carried out. In addition, establishing an immunotherapy cohort of CRC patients is crucial for gaining a deeper understanding of how HDAC10 influences immunotherapy efficacy and the underlying biological regulatory mechanisms involved.

In conclusion, this study highlights that HDAC10 promotes cancer growth and metastasis while inhibiting TME and impairing immunotherapy efficacy. Furthermore, HDAC10 has been identified as an independent prognostic factor in CRC patients. Inhibiting HDAC10 expression emerges as a promising therapeutic strategy for cancer treatment and enhancing anti-tumor immune responses.

CONCLUSION

In summary, we identified a novel biomarker in CRC. Through analysis of human CRC samples and biological investigations, we found that HDAC10 functions as a putative oncogene, driving CRC cell proliferation and migration in vitro. Moreover, HDAC10 has the capacity to modulate TME and subsequently impact immunotherapy efficacy. Given these significant findings, targeting the regulation of HDAC10 expression or function emerges as a promising therapeutic approach for the treatment of CRC.