M6A and lactylation modification-related signature predicts prognosis and immunotherapy response in hepatocellular carcinoma

doi:10.21203/rs.3.rs-4950216/v1

Download PDF

Article

M6A and lactylation modification-related signature predicts prognosis and immunotherapy response in hepatocellular carcinoma

https://doi.org/10.21203/rs.3.rs-4950216/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

This study aims to investigate the role of lactylation and m6A modification-related genes in the tumor microenvironment and immunotherapy for hepatocellular carcinoma (HCC) patients. RNA-sequence data and corresponding clinical information of HCC were obtained from the TCGA and ICGC datasets. LASSO Cox regression analysis was implied to construct a lactylation-m6A related prognostic model. The 7-gene signature was established and effectively stratified patients into high- and low-risk groups. Further analysis revealed significant differences between the two risk groups in terms of tumor microenvironment, expression levels of immune checkpoint genes, and drug responsiveness. Specifically, the high-risk group exhibited increased immune cell infiltration, lower IC50 values for several drugs including 5-fluorouracil, afatinib, crizotinib, cediranib, taselisib, and staurosporine; Whereas the low-risk group displayed reduced stromal component proportions and better responses to entinostat, irinotecan, KRAS inhibitors, cisplatin, axitinib, and topotecan. Functionally, knockdown of TCOF1 and HDAC1 significantly attenuated the migration and invasive capacity of Huh-7 liver cancer cells. The lactylation-m6A related prognostic model exhibited robust predictive efficiency in HCC. TCOF1 and HDAC1 may be promising tumor biomarkers for HCC and more researches are needed to validate these results.

Biological sciences/Cancer

Biological sciences/Genetics

lactylation

m6A

hepatocellular carcinoma

prognosis

immunotherapy

Hepatocellular carcinoma (HCC) ranks among the most widespread malignancies worldwide, serving as the third leading cause of cancer-related fatalities. It is widely accepted that HCC typically arises as a result of prolonged liver pathologies^{[1, 2]}, such as nonalcoholic fatty liver disease (NAFLD), or chronic infections with either the Hepatitis B virus (HBV) or the hepatitis C virus (HCV). Unfortunately, due to the absence of specific symptoms in HCC’s early stages, 85% of patients present were diagnosed at an advanced stage and distanted metastases, which was a key factor contributing to the disease's generally bleak prognosis^{[3, 4]}. Despite the progress has been made in the molecular mechanism of pathogenesis and drug therapy, there is still challenging to deliver effective therapy due to tumor heterogeneity, metabolic reprogramming, tumor recurrence and drug resistance^[5]. Therefore, it is necessary to explore the tumor microenvironment and identify more precise prognostic biomarkers for HCC patients, enhancing personalized treatment strategies and improving patient outcomes. The recent discovery of lactylation, an epigenetic modification involving the addition of a lactyl group to lysine residues in chromatin, has expanded our understanding of the complex mechanisms regulating gene transcription^[6]. This novel modification has been shown to directly stimulate gene expression.

Subsequent studies have confirmed that there are currently two modification pathways for lactylation. One of the pathways is that the L-lactate forms lactoyl-CoA mediated lactylation by the catalysis of P300^[6]. Lactoylglutathione (LGSH), formed by methylglyoxal, can also mediate the lactylation of proteins without enzyme catalysis. Under the action of glyxoalase2 (GLO2), LGSH can be disassemble into glutathione and D-lactate^[7]. Indeed, lactylation represents a fascinating intersection between lactate metabolism and epigenetic regulation, offering new insights into the interplay between cellular metabolism and gene expression control^[8]. Recently, lactylation has emerged as an important event that modulates tumorigenesis^[9], tumor progression^[10], metabolism of tumor cells^[11], and tumoral immune response. For example, research found that histone lactylation could drive the tumor progression. While this phenomenon could be reversed by activating the Hippo pathway^[12]. Hypoxia-inducible factor (HIF)-1α lactylation promoted angiogenesis and vasculogenic mimicry in prostate cancer^[13].

N6-methyladenosine (m6A) modifications represent the most prevalent type of RNA modifications, exerting significant influence on various aspects of RNA biology, including nuclear processing, translation efficiency, and mRNA stability^[14]. These modifications play a crucial role in the regulation of gene expression and signal transduction. The m6A methylation system is composed of three main components: methyltransferase (Writers), demethylase (Erasers) and m6A binding proteins (Readers)^[14]. Abnormalities in the m6A RNA modification machinery have been implicated in various diseases, particularly cancer, where they contribute to tumor progression by affecting cell proliferation, differentiation, migration, and apoptosis^[15]. Recent findings suggest that lactylation and m6A modifications may interact and jointly regulate pathological processes. In the development of idiopathic pulmonary fibrosis (IPF), crosstalk between AECs and myofibroblasts via lactylation and m6A is thought to accelerated the progression of IPF^[16]. Furthermore, METTL16 lactylation at site K229 is induced by cuproptosis in development of gastric cancer^[17]. In tumor immune microenvironment, Lactate accumulated potently induced METTL3 upregulation via H3K18 lactylation, and the upregulated METTL3 mediated m6A modification, which is responsible for immunosuppressive capacity of tumor-infiltrating myeloid cells^[18]. In conclusion, the crosstalk between lactylation and m6A modifications represents an emerging area of research with potential implications for our understanding of complex disease processes like tumor. Revealing the molecular mechanisms underlying two epigenetic modifications may uncover novel therapeutic targets. However, the crosstalk histone lactylation and m6A modification in the development of HCC is still unclear.

With the rapid development of bioinformatics and the establishment of the Cancer Genome Atlas (TCGA), genome, proteome, transcriptome, and metabolome are explored to the tumorigenesis mechanism and biomarkers for the prognosis and therapy of cancer. To help understand the role of histone lactylation and m6A in HCC development, we analyzed the expression of lactylation-m6A related genes and constructed a prognostic signature for HCC patients in TCGA cohort. We further analyzed prognostic signature in tumor immune microenvironment and drug prediction. Our study may provide more data for the prognostic biomarkers and molecular mechanisms in HCC.

2.1 Data collection

The RNA-sequence data, somatic mutation data, copy number variation (CNV) data, and clinicopathological data pertaining to HCC patients were extracted from The Cancer Genome Atlas (TCGA; https://portal.gdc.cancer.gov/repository). The clinicopathological information of HCC patients within the TCGA dataset was detailed in Supplementary Table 1. The lactylation-related genes were collected from previous studies^[19]. An external validation dataset consisting of gene expression data for 231 HCC patients was acquired from The International Cancer Genome Consortium (ICGC) (https://docs.icgc-argo.org/docs/data-access/icgc-25k-data). The normalization of gene expression data was executed using the Perl programming language (http://www.perl.org/). Data analysis was performed with R software (version 4.3.1) and R Bioconductor packages.

2.2 Identification of correlation, differential expressed, and prognosis related genes

Pearson correlation analysis was utilized to discern crosstalk genes between lactylation-relevant genes and a set of 23 m6A-linked genes, adopting rigorous selection criteria of correlation magnitude > 0.5 and statistical significance at P < 0.01. Lactylation modification-related DEGs were identified on basis of the following criteria: false discovery rate (FDR) < 0.05 and a log2 fold change > 1. Subsequently, the univariate Cox analysis was performed to confirm the prognostic intersection genes (IGs) (P < 0.05). A protein-protein interaction (PPI) network was then assembled using the STRING online resource (http://string-db.org/cgi/), with edges exhibiting an interaction score surpassing 0.2 deemed statistically significant and incorporated into the network representation.

2.3 Identification of HCC subtypes

HCC patients in TCGA dataset were divided into different subtypes through an unsupervised methodology by the ConsenSuClusterPlus R package. Principal component analysis (PCA) was used to identify the ability of the clusters to discriminate between subtypes. The differences over survival rate were compared between subtypes by K-M analysis.

2.4 Construction and validation of a lactylation-m6A related prognostic signature

The construction of a lactylation-m6A-informed prognostic signature involved genes significantly associated with survival (P < 0.01) into a LASSO Cox regression model, facilitated by the "glmnet" R package. The TCGA dataset was designated as the training dataset. The risk score calculation followed the equation: RiskScore = Σ (risk genes’ expression × coefficient). Patients in the TCGA dataset were then categorized into high- and low-risk groups according to the median risk score. Differential survival rates between the two risk groups were examined via Kaplan-Meier (K-M) analysis. PCA and T-distributed stochastic neighbor embedding (tSNE) were adopted to validate the ability of the prognostic signature to discriminate HCC patients in different risk groups. The predictive ability of the prognostic model was evaluated by the receiver operating curves (ROC) curves analysis using the R package “timeROC”.

Validation of the prognostic signature was carried out using data derived from the ICGC cohort. Univariate and multivariate Cox regression analyses were performed to ascertain the independence of the risk score as a prognostic factor by evaluating the hazard ratio (HR), 95% confidence interval (CI), and P-values.

2.5 Enrichment analysis

Gene Set Enrichment Analysis (GSEA) was used to analyze the remarkable enrichment pathways between high- and low- risk groups via the R package “clusterProfiler”, “org.Hs.eg.db”, and “enrichplot”. P < 0.05 was regarded as statistically enriched.

2.6 Immune cell infiltration, tumor mutation burden and tumor microenvironment analysis

Calculation of the tumor mutation burden (TMB) was performed according to the equation: TMB = (total mutation/total covered bases) ×10⁶. A comparison of TMB levels between the high- and low-risk groups was undertaken using TCGA cohort data, with the "maftools" package facilitating the examination of mutation data in each risk category. To evaluate the association between risk score and tumor-infiltrating immune cells, seven algorithms – XCELL, TIMER, QUANTISEQ, MCPCOUNTER, EPIC, CIBERSORT-ABS, and CIBERSORT – were applied to calculate Spearman correlations. The ESTIMATE algorithm was utilized to determine the stromal score, immune score, and estimate score.

2.7 The prediction of drug sensitivity

The R package "oncoPredict" was utilized to predict differences in drug sensitivity between patients classified as high- and low-risk.

2.8 Cell culture and transfection

The human HCC Huh-7 cell line was obtained from the China Center for Type Culture Collection (Wuhan, China). The cell line was cultured in medium DMEM (Hyclone, USA) supplemented with 10% fetal bovine serum (Siji Green, China) and 1% penicillin-streptomycin (Gibco, USA) at 37℃ in a humidified incubator with 5% CO2. The TCOF1 and HDAC1 siRNAs were synthesized by Ruibo Biotechnology (Guangzhou, China). Those siRNA were transfected into Huh-7 cell using Lipofectamine™ 3000.

2.9 Wound healing and Transwell® assays

The wound healing assay was conducted to assess cell migratory capacity. Briefly, cells in 6-well plates, approaching approximately 90% fusion, were scratched with a sterile 10 µL pipette tip to create a central wound in each well. Following a 24-hour incubation, photographs were taken using an inverted microscope to determine gap in at least three randomly selected fields. The migration distance was determined by Image J software. Cell invasion ability was measured by Transwell® assays using 24-well Transwell® chambers with 8 µm polycarbonate membranes (Millipore, MA, USA) and was performed as previously described^[20].

2.10 Statistical Analysis

All data were explored and visualized with the use of the R 4.3.1 software. For each analysis, P < 0.05 was considered statistically significant, *P < 0.05, **P < 0.01, ***P < 0.001.

3.1 Analysis of lactylation-m6A related genes and DEGs in HCC patients

By correlation analysis, we identified 207 crosstalk genes between and lactylation and m6A, and the co-expression network was shown in Fig. 1A. We identified the lactylation-related DEGs between normal and HCC tissues which included 125 upregulated genes and 1 downregulated gene (Fig. 1B). Then, 87 intersection genes (IGs) were identified from 207 lactylation-m6A related genes and 126 DEGs (Fig. 1C). PPI networks were constructed for IGs (Fig. 1D). To further explore the function of m6A-lactylation related genes in HCC, GO functional enrichment analysis based on IGs was implemented. The biological processes (BP) were involved in RNA export from nucleus, RNA localization, nucleic acid transport, RNA transport, and establishment of RNA localization (Fig. 1E). The cellular components (CC) included chromosomal region, nuclear speck, chromosome, centromeric region, and condensed chromosome. The molecular functions (MF) included tubulin binding, chromatin DNA binding, and single-stranded RNA binding. According to Fig. 1F, these intersection genes were mutated in 12.64% of samples, and PRKDC exhibited the highest mutation frequency (6%).

3.2 Identification of HCC subtypes based on prognostic IGs

Univariate Cox regression analysis was employed to identify 67 prognostic IGs (Fig. 2A). Then HCC patients in TCGA cohort were separated into two clusters according to the expression of prognostic IGs (Fig. 2B, and Fig. S1), resulting in 293 cases assigned to cluster-1, and 50 cases of cluster-2. The PCA performed in all HCC samples revealed that there was little overlap between the two clusters (Fig. 2C). Survival analysis demonstrated that patients in cluster-1 had a significantly decreased survival rate compared to those in cluster-2 (Fig. 2D). A Sankey diagram was employed to illustrate the association between clusters and corresponding clinical traits (Fig. 2E). Infiltration indicated that there were differences in naive B cell, memory B cell, CD4 T cells M0/M1 macrophages and neutrophils between two clusters (Fig. 2F).

3.3 Constructing a lactylation and m6A related prognostic model

The 67 prognosis-related IGs were subjected to lasso regression analysis in TCGA cohort. According to the minimum criteria, a risk signature based on 7 prognosis-related IGs (G6PD, PRKDC, HDAC1, CCT5, KIF2C, TCOF1 and PPM1G) was to establish (Fig. 3A-B). Risk score per patient = TCOF1*(0.013) + HDAC1*(0.003) + PPM1G*(0.013) + CCT5*(0.003) + KIF2C*(0.024) + PRKDC*(0.003) + G6PD*(0.005). Based on their risk scores relative to the median value, patients were classified into low- or high-risk groups. K-M survival curves demonstrated that patients in the high-risk group exhibited significantly lower survival rates compared to their low-risk counterparts (Fig. 3C). The prognostic model's predictive capability was evidenced by ROC values, which showed its success in predicting HCC patient survival in the TCGA cohort (1-year AUC = 0.791, 3-year AUC = 0.699, and 5-year AUC = 0.649) (Fig. 3D). PCA and t-SNE analyses further verified this prognostic model separated the high-risk patients from low-risk group in TCGA cohort (Fig. 3E-F). Risk survival status plots revealed a negative association between risk scores and patient survival, as higher scores correlated with shorter overall survival in TCGA cohort (Fig. 3G-H). Furthermore, a heatmap highlighted the elevated expression of the risk genes in the high-risk group, implicating their role in the poorer prognosis observed in these patients (Fig. 3I). Regardless of age (Fig. S2A-B), tumor grade (Fig. S2C-D), gender (Fig. S2E-F), or stage (Fig. S2G-H), patients in the high-risk group consistently experienced more reductions in survival rates.

Data from ICGC was utilized to verify the accuracy and generalizability of this risk signature. Firstly, patients in ICGC cohort were stratified into high- and low-risk groups using the same equation applied in TCGA analysis. Consistent with the TCGA findings, patients with high-risk score in ICGC cohort displayed significantly shorter survival times (Fig. S3A). According to PCA and t-SNE analysis, these 7 risk genes could be assigned to different risk groups of HCC patients in ICGC cohort (Fig. S3B-C). The distribution of the risk score, survival status of each patient and expression of these 7 risk genes were exhibited in Fig. S3D-F.

To investigate the independence of the risk score as a predictor of HCC patient prognosis, univariate and multivariate Cox regression analyses were conducted for the TCGA and ICGC cohorts. As shown in Fig. S4A-B, that the risk score was a statistically significant independent predictor of patient survival in TCGA cohort (for univariate Cox analysis, risk score, HR = 1.448, 95% CI = 1.315–1.595, P < 0.001; for multivariate Cox analysis, HR = 1.405, 95% CI = 1.258–1.568, P < 0.001). As similar with the results in TCGA cohort, the risk score was also proved to be an independent factor in ICGC cohort (Fig. S4C-D). In summary, these results demonstrated our risk signature might be an independent factor to predict the prognosis of HCC patients.

3.4 The correlation of TMB and prognostic signature

We investigated the copy number variation (CNV) in the seven risk genes across all HCC samples. As shown in Fig. 4A, while KIF2C and HDAC1 showed a decrease in copy number, the remaining five risk genes all displayed increased copy numbers. It had been reported that TMB was a biomarker that indicated response to immunotherapy. Hence, we procured mutation data for HCC patients from the TCGA database and computed TMB values for each subject to explore the association between TMB and our prognostic signature. The findings revealed that the top 20 driver gene with frequency of alterations significantly differed between high- and low-risk groups (Fig. 4B-C). Moreover, patients with high TMB values had a poorer prognosis compared to those with low TMB (Fig. 4D). Importantly, HCC patients who had the high-TMB combined with a high-risk score patients had a worst prognosis (Fig. 4E). These results indicated that the prognostic signature could reflect the genomic stability for HCC patients.

3.5 Correlation analysis between the risk score and tumor microenvironment

To investigate the biological differences between the high- and low-risk groups, Gene Set Enrichment Analysis (GSEA) was conducted on the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways for each group. The top ten enriched pathways in high-risk group were predominantly correlated with cell cycle, DNA replication, ECM receptor interaction, and intestinal immune network (Fig. 5A). While, the low-risk group was highly enrichment in drug metabolism, fatty acid metabolism and other metabolic processes (Fig. 5B). Furthermore, we observed a positive correlation between the risk score and the abundance of most immune cell types (Fig. 5C), highlighting the potential impact of the immune microenvironment on the risk stratification. Indeed, the high-risk group displayed a higher immune score and a lower stromal score compared to the low-risk group (Fig. 5D). Given the critical role of immune checkpoints in dictating responses to immunotherapy, we analyzed the expression of immune checkpoints related genes in high- and low-risk groups. Multiple costimulatory molecules, such as CD70, TNFRSF4/9/14, IFNG, IL-1B, IL-2RA, and HMGB1, were upregulated in high-risk group (Fig. 5E). Some coinhibitory molecules, such as VEGFA, SLAMF7, CTLA4, IL-10, LAG3 and EDNRB were showed a higher in high-risk group (Fig. 5F). These results proved that our risk score possibly predict the immunotherapy for HCC patients.

3.6 Prediction drug sensitivity

In consideration of the GSVA analysis showed that the low-risk genes were enriched in drug metabolism, we proceeded to compare drug sensitivity between the high- and low-risk groups by IC50. As showed in Fig. 6, some drugs such as 5-fluorouracil, afatinib, crizotinib, cediranib, taselisib and stauroporine were more sensitive for patients in high-risk group. While entinostat, irinotecan, KRAS inhibitor, cisplatin, axitinib and topotecan were more sensitive to the treatment for patients in low-risk group. Those data suggested that the risk score may predict the patients' response to treatment with certain drugs.

3.7 The protein expression of risk genes

The protein levels of these 7 risk genes were explored in the Human Protein Atlas (HPA) database. As shown in Fig. 7A, the expression of TCOF1, HDAC1 and PRKDC was upregulated in HCC tissue than normal tissue. While, the expression of G6PD, CCT5, KIF2 and PPM1G were very low or no detected in HPA database. In addition, we found that the expression of PRKDC was no correlation with the prognosis of patients. Further analysis showed that TCOF1 (Fig. 7B) and HDAC1 (Fig. 7C) was positive with M0 macrophages, Tregs, and memory B cells, while negative with M1/M2 macrophages, neutrophils and monocytes.

3.8 TCOF1- and HDAC1-silencing inhibited the cell migration and invasion

In order to further explore the role of TCOF1 and HDAC1 in HCC development, we used the siRNA to silence TCOF1 and HDAC1 in Huh-7 cell, respectively (Fig. 8A). As shown in Fig. 8B, cell invasion ability decreased significantly by silencing TCOF1 and HDAC1. Similarly, wound healing analysis also verified that silencing TCOF1 and HDAC1 inhibited the migration of Huh-7 cell (Fig. 8C). Our results demonstrate that TCOF1 and HDAC1 influence migration and invasion of Huh-7 cell.

Lactate, an end-product of intracellular glycolysis, has evolved from being a metabolic waste product to a molecule with recognized significance in both physiological and pathological processes, particularly in the context of cancer. This paradigm shift was sparked by the discovery of the Warburg effect, which highlights the aberrant reliance of tumor cells on aerobic glycolysis for energy production, leading to excessive lactate accumulation^[21]. Lactate has since been implicated in two core characteristics of cancer: immune evasion and metabolic reprogramming^[22]. Recent studies have exhibited that lactate can mediate protein modification, regulate genes expression in macrophages, somatic cells, and cancer cells by lysine lactylation^[23]. As one of the most ubiquitous types of posttranscriptional modification, m6A is reported to involve in the development and progression of numerous diseases, including cancer. However, the interplay between m6A and lactylation remains unexplored. To our knowledge, this investigation represents the first attempt to examine the relationship between m6A-lactylation-related genes and prognosis in HCC patients. Our findings may provide novel insights into the role of these modifications in HCC progression.

In our study, we identified the prognostic lactylation-m6A related genes and HCC patients in TCGA data cohort were divided into two cluster. Patients in cluster-1 had a poor prognosis than these in cluster-2. The immune cells infiltration was different between two cluster. Subsequently, employing LASSO regression, a prognostic model was formulated and patients were stratified into high- and low-risk categories. Through univariate and multivariate Cox regression analyses, this model encompassing 7 risk-associated genes was established as an independent prognostic determinant. The K-M survival analysis revealed significantly inferior survival among high-risk patients relative to their low-risk counterparts. Similar findings were replicated in the ICGC dataset. Notably, accumulating evidence underscores the predictive relevance of TMB, not only in the context of immunotherapy, but also when combined with chemotherapy^{[24, 25]}. In our research, we found that patients who had the high-TMB combined with a high-risk score had a poorer prognosis. GSEA analysis was used to explore the different pathways with enrichment in high- and low-risk groups. High-risk group were highly correlated with DNA replication, ECM receptor interaction, intestinal immune network, and low-risk group was highly enrichment in drug metabolism, fatty acid metabolism and other metabolic processes. Furthermore, we found that there was a higher immune score, lower stromal score, and higher expression of coinhibitory molecules in high-risk group compared with low-risk group. The drug response in high and low-risk group also showed significant difference which indicated the risk score may successfully predict the patients' response to treatment with certain drugs.

We further investigated the protein expression of risk genes (G6PD, PRKDC, HDAC1, CCT5, KIF2C, TCOF1 and PPM1G) in HAP database. Consistent with the expression of RNA level, the protein expression of TCOF1, HDAC1 and PRKDC was upregulated in HCC tissue than normal tissue. While, the protein of G6PD, CCT5, KIF2 and PPM1G were very low or no detected in in HPA database which probably be due to post-translation modification or degradation of protein. K-M survival analysis showed the high expression of TCOF1 and HDAC1 was correlated with a shorter survival time, but PRKDC had no influence for patients' survival. TCOF1 is generally regarded as a nucleolar factor that collaborates with upstream binding factor (UBF) to regulate ribosomal DNA (rDNA) transcription in eukaryotes^[26]. It has been established that TCOF1 exerts a significant influence on both neural crest development and the biosynthesis of ribosomes^{[27, 28]}. However, recent studies have found that TCOF1 probably appears to be antitumorigenic or protumorigenic depending on the tumor type and conditions. One research reported that TCOF1 downregulation was observed in human T-cell acute lymphoblastic leukemia (T-ALL) cells and TCOF1 haploinsufficiency promoted an increased incidence of T-ALL in mice^[29]. Conversely, upregulation of TCOF1 is observed in triple-negative breast cancer (TNBC)^[30], HCC^{[31, 32]} and correlates with a poor prognosis. Mechanically, TCOF1 regulates KRAS-activated genes, epithelial‐mesenchymal transition (EMT) genes, and promotes the expression ribosomal RNA (rRNA) production which is a hallmark of cancer^[32]. Histone deacetylase 1 (HDAC1) is one of the classes Ⅰ HDACs, and impacts whole animal physiology and health by not only directly regulating transcription via deacetylation of histones but also enzyme through activity direct deacetylation^[33]. In T cell-mediated immune diseases, HDAC1 acted as the key regulator to acute and chronic adaptation to environment stimuli such as allergen, stress^{[34, 35]}. However, the regulatory function of HDAC1 on inflammatory signaling is paradoxical, as it can either stimulate or inhibit such responses, contingent upon the specific contextual cues or extracellular stimuli encountered^[33]. Similarity, HDAC1 also have complex role in regulating cell proliferation, such as an acceleration for cancer cells^[36] and an anti-proliferative effect for endothelial cells. HDAC1 was identified as the critical direct m6A target in bipotent progenitor^[37]. Overexpression of HDAC1 was supposed to be associated with tumor progression and metastasis^{[38, 39]}. HDAC1 was also involved in drug resistance by regulating the genes expression^[40]. Several HDAC inhibitors have exhibited promising results in the treatment for multiple carcinoma^[41]. In our study, we identified that TCOF1 and HDAC1 were the key prognostic genes for HCC patients and more research is needed to further reveal the role of these two genes in HCC progression.

However, there is still some limitations in our study. The first limitation is that protein lactylation is a novel field, more and more lactylation-related genes will be discovered in the future. Secondly, our reliance on predominantly retrospective data sourced from public repositories necessitates the need for further corroboration through fundamental experimental validations. Thirdly, our work lacks a comprehensive elucidating of the precise molecular mechanisms controlling the critical roles of key genes in HCC progression. Finally, the complex interaction between HCC cells and the immune microenvironment in the context of lactation-M6A modification remains an area for further study.

In summary, our research might provide a novel perspective for forecasting the prognosis of HCC patients and uncovered the relationship between lactylation-related genes and HCC. Our results provided great potential for advancing research of lactylation and m6A in HCC.

Author Contributions statement

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Huili Ren, Jianhua Liu and Shaohui Zhang. The first draft of the manuscript was written by Huili Ren and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Fundings

This work was supported by National Natural Science Foundation of China grants (82404019).

Data Availability

The data are available in TCGA (https://portal.gdc.cancer.gov/) and ICGG (https://docs.icgc-argo.org/docs/data-access/icgc-25k-data).

Declarations

Conflict of interest: The authors have no relevant financial or non-financial interests to disclose.

Ethical approval: Not applicable.

Consent to participate: Not applicable.

J. D. Yang, P. Hainaut, G. J. Gores, A. Amadou, A. Plymoth, and L. R. Roberts. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol, 2019, 16(10): 589-604.
M. G. Refolo, C. Messa, V. Guerra, B. I. Carr, and R. D'Alessandro. Inflammatory Mechanisms of HCC Development. Cancers (Basel), 2020, 12(3)
W. Wang and C. Wei. Advances in the early diagnosis of hepatocellular carcinoma. Genes Dis, 2020, 7(3): 308-319.
Y. Jiang, A. Sun, Y. Zhao, W. Ying, H. Sun, X. Yang, B. Xing, W. Sun, L. Ren, B. Hu, C. Li, L. Zhang, G. Qin, M. Zhang, N. Chen, M. Zhang, Y. Huang, J. Zhou, Y. Zhao, M. Liu, X. Zhu, Y. Qiu, Y. Sun, C. Huang, M. Yan, M. Wang, W. Liu, F. Tian, H. Xu, J. Zhou, Z. Wu, T. Shi, W. Zhu, J. Qin, L. Xie, J. Fan, X. Qian, F. He, and Consortium Chinese Human Proteome Project. Proteomics identifies new therapeutic targets of early-stage hepatocellular carcinoma. Nature, 2019, 567(7747): 257-261.
E. Ramon-Gil, D. Geh, and J. Leslie. Harnessing neutrophil plasticity for HCC immunotherapy. Essays Biochem, 2023, 67(6): 941-955.
D. Zhang, Z. Tang, H. Huang, G. Zhou, C. Cui, Y. Weng, W. Liu, S. Kim, S. Lee, M. Perez-Neut, J. Ding, D. Czyz, R. Hu, Z. Ye, M. He, Y. G. Zheng, H. A. Shuman, L. Dai, B. Ren, R. G. Roeder, L. Becker, and Y. Zhao. Metabolic regulation of gene expression by histone lactylation. Nature, 2019, 574(7779): 575-580.
D. O. Gaffney, E. Q. Jennings, C. C. Anderson, J. O. Marentette, T. Shi, A. M. Schou Oxvig, M. D. Streeter, M. Johannsen, D. A. Spiegel, E. Chapman, J. R. Roede, and J. J. Galligan. Non-enzymatic Lysine Lactoylation of Glycolytic Enzymes. Cell Chem Biol, 2020, 27(2): 206-213 e206.
J. H. Wang, L. Mao, J. Wang, X. Zhang, M. Wu, Q. Wen, and S. C. Yu. Beyond metabolic waste: lysine lactylation and its potential roles in cancer progression and cell fate determination. Cell Oncol (Dordr), 2023, 46(3): 465-480.
E. Kotsiliti. Lactylation and HCC progression. Nat Rev Gastroenterol Hepatol, 2023, 20(3): 131.
Y. He, Z. Ji, Y. Gong, L. Fan, P. Xu, X. Chen, J. Miao, K. Zhang, W. Zhang, P. Ma, H. Zhao, C. Cheng, D. Wang, J. Wang, N. Jing, K. Liu, P. Zhang, B. Dong, G. Zhuang, Y. Fu, W. Xue, W. Q. Gao, and H. H. Zhu. Numb/Parkin-directed mitochondrial fitness governs cancer cell fate via metabolic regulation of histone lactylation. Cell Rep, 2023, 42(2): 112033.
J. Su, Z. Zheng, C. Bian, S. Chang, J. Bao, H. Yu, Y. Xin, and X. Jiang. Functions and mechanisms of lactylation in carcinogenesis and immunosuppression. Front Immunol, 2023, 14: 1253064.
B. Xie, J. Lin, X. Chen, X. Zhou, Y. Zhang, M. Fan, J. Xiang, N. He, Z. Hu, and F. Wang. CircXRN2 suppresses tumor progression driven by histone lactylation through activating the Hippo pathway in human bladder cancer. Mol Cancer, 2023, 22(1): 151.
Y. Luo, Z. Yang, Y. Yu, and P. Zhang. HIF1alpha lactylation enhances KIAA1199 transcription to promote angiogenesis and vasculogenic mimicry in prostate cancer. Int J Biol Macromol, 2022, 222(Pt B): 2225-2243.
D. Wang, Y. Zhang, Q. Li, A. Zhang, J. Xu, Y. Li, W. Li, L. Tang, F. Yang, and J. Meng. N6-methyladenosine (m6A) in cancer therapeutic resistance: Potential mechanisms and clinical implications. Biomed Pharmacother, 2023, 167: 115477.
Z. M. Zhu, F. C. Huo, J. Zhang, H. J. Shan, and D. S. Pei. Crosstalk between m6A modification and alternative splicing during cancer progression. Clin Transl Med, 2023, 13(10): e1460.
P. Wang, D. Xie, T. Xiao, C. Cheng, D. Wang, J. Sun, M. Wu, Y. Yang, A. Zhang, and Q. Liu. H3K18 lactylation promotes the progression of arsenite-related idiopathic pulmonary fibrosis via YTHDF1/m6A/NREP. J Hazard Mater, 2024, 461: 132582.
L. Sun, Y. Zhang, B. Yang, S. Sun, P. Zhang, Z. Luo, T. Feng, Z. Cui, T. Zhu, Y. Li, Z. Qiu, G. Fan, and C. Huang. Lactylation of METTL16 promotes cuproptosis via m(6)A-modification on FDX1 mRNA in gastric cancer. Nat Commun, 2023, 14(1): 6523.
J. Xiong, J. He, J. Zhu, J. Pan, W. Liao, H. Ye, H. Wang, Y. Song, Y. Du, B. Cui, M. Xue, W. Zheng, X. Kong, K. Jiang, K. Ding, L. Lai, and Q. Wang. Lactylation-driven METTL3-mediated RNA m(6)A modification promotes immunosuppression of tumor-infiltrating myeloid cells. Mol Cell, 2022, 82(9): 1660-1677 e1610.
Z. Cheng, H. Huang, M. Li, X. Liang, Y. Tan, and Y. Chen. Lactylation-Related Gene Signature Effectively Predicts Prognosis and Treatment Responsiveness in Hepatocellular Carcinoma. Pharmaceuticals (Basel), 2023, 16(5)
H. Ren, Y. Chen, Z. Ao, Q. Cheng, X. Yang, H. Tao, L. Zhao, A. Shen, P. Li, and Q. Fu. PDE4D binds and interacts with YAP to cooperatively promote HCC progression. Cancer Lett, 2022, 541: 215749.
M. G. Vander Heiden, L. C. Cantley, and C. B. Thompson. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science, 2009, 324(5930): 1029-1033.
L. Chen, L. Huang, Y. Gu, W. Cang, P. Sun, and Y. Xiang. Lactate-Lactylation Hands between Metabolic Reprogramming and Immunosuppression. Int J Mol Sci, 2022, 23(19)
W. Sun, M. Jia, Y. Feng, and X. Cheng. Lactate is a bridge linking glycolysis and autophagy through lactylation. Autophagy, 2023, 19(12): 3240-3241.
T. Fulton-Ward and G. Middleton. The impact of genomic context on outcomes of solid cancer patients treated with genotype-matched targeted therapies: a comprehensive review. Ann Oncol, 2023
Y. Huang, Y. F. Chau, H. Bai, X. Wu, and J. Duan. Biomarkers for Immunotherapy in Driver-Gene-Negative Advanced NSCLC. Int J Mol Sci, 2023, 24(19)
B. C. Valdez, D. Henning, R. B. So, J. Dixon, and M. J. Dixon. The Treacher Collins syndrome (TCOF1) gene product is involved in ribosomal DNA gene transcription by interacting with upstream binding factor. Proc Natl Acad Sci U S A, 2004, 101(29): 10709-10714.
A. Werner, S. Iwasaki, C. A. McGourty, S. Medina-Ruiz, N. Teerikorpi, I. Fedrigo, N. T. Ingolia, and M. Rape. Cell-fate determination by ubiquitin-dependent regulation of translation. Nature, 2015, 525(7570): 523-527.
B. Gonzales, D. Henning, R. B. So, J. Dixon, M. J. Dixon, and B. C. Valdez. The Treacher Collins syndrome (TCOF1) gene product is involved in pre-rRNA methylation. Hum Mol Genet, 2005, 14(14): 2035-2043.
Z. Wen, R. Finn, X. Gao, L. Li, A. Hebert, E. A. Ranheim, Y. Zhou, G. Yun, J. P. Roose, J. J. Coon, R. Shiang, R. Wen, M. Yu, D. Wang, and J. Zhang. Tcof1 haploinsufficiency promotes early T cell precursor-like leukemia in Nras(Q61R/+) mice. Leukemia, 2022, 36(4): 1167-1170.
J. Hu, Y. Lai, H. Huang, S. Ramakrishnan, Y. Pan, V. W. S. Ma, W. Cheuk, G. Y. K. So, Q. He, C. Geoffrey Lau, L. Zhang, W. C. S. Cho, K. M. Chan, X. Wang, and Y. Rebecca Chin. TCOF1 upregulation in triple-negative breast cancer promotes stemness and tumour growth and correlates with poor prognosis. Br J Cancer, 2022, 126(1): 57-71.
Y. Zhang, N. Li, L. Yang, W. Jia, Z. Li, Q. Shao, and X. Zhan. Quantitative phosphoproteomics reveals molecular pathway network alterations in human early-stage primary hepatic carcinomas: potential for 3P medical approach. EPMA J, 2023, 14(3): 477-502.
C. Wu, D. Xia, D. Wang, S. Wang, Z. Sun, B. Xu, and D. Zhang. TCOF1 coordinates oncogenic activation and rRNA production and promotes tumorigenesis in HCC. Cancer Sci, 2022, 113(2): 553-564.
L. S. Dunaway and J. S. Pollock. HDAC1: an environmental sensor regulating endothelial function. Cardiovasc Res, 2022, 118(8): 1885-1903.
L. Goschl, T. Preglej, N. Boucheron, V. Saferding, L. Muller, A. Platzer, K. Hirahara, H. Y. Shih, J. Backlund, P. Matthias, B. Niederreiter, A. Hladik, M. Kugler, G. A. Gualdoni, C. Scheinecker, S. Knapp, C. Seiser, R. Holmdahl, K. Tillmann, R. Plasenzotti, B. Podesser, D. Aletaha, J. S. Smolen, T. Karonitsch, G. Steiner, W. Ellmeier, and M. Bonelli. Histone deacetylase 1 (HDAC1): A key player of T cell-mediated arthritis. J Autoimmun, 2020, 108: 102379.
P. Ma and R. M. Schultz. HDAC1 and HDAC2 in mouse oocytes and preimplantation embryos: Specificity versus compensation. Cell Death Differ, 2016, 23(7): 1119-1127.
S. Yoon and G. H. Eom. HDAC and HDAC Inhibitor: From Cancer to Cardiovascular Diseases. Chonnam Med J, 2016, 52(1): 1-11.
J. Sun, Y. Wang, H. Fu, F. Kang, J. Song, M. Xu, G. Ning, J. Wang, W. Wang, and Q. Wang. METTL3-mediated m6A Methylation Controls Pancreatic Bipotent Progenitor Fate and Islet Formation. Diabetes, 2023
Y. Li and E. Seto. HDACs and HDAC Inhibitors in Cancer Development and Therapy. Cold Spring Harb Perspect Med, 2016, 6(10)
R. C. Ramaker, A. A. Hardigan, E. R. Gordon, C. A. Wright, R. M. Myers, and S. J. Cooper. Pooled CRISPR screening in pancreatic cancer cells implicates co-repressor complexes as a cause of multiple drug resistance via regulation of epithelial-to-mesenchymal transition. BMC Cancer, 2021, 21(1): 632.
R. Hai, L. He, G. Shu, and G. Yin. Characterization of Histone Deacetylase Mechanisms in Cancer Development. Front Oncol, 2021, 11: 700947.
R. Parveen, D. Harihar, and B. P. Chatterji. Recent histone deacetylase inhibitors in cancer therapy. Cancer, 2023, 129(21): 3372-3380.

No competing interests reported.

SupplementaryMaterial.docx

Download PDF

Reviews received at journal
30 Oct, 2024
Reviewers agreed at journal
30 Oct, 2024
Reviewers invited by journal
19 Sep, 2024
Editor assigned by journal
13 Sep, 2024
Editor invited by journal
06 Sep, 2024
Submission checks completed at journal
04 Sep, 2024
First submitted to journal
21 Aug, 2024

You are reading this latest preprint version

M6A and lactylation modification-related signature predicts prognosis and immunotherapy response in hepatocellular carcinoma

Status:

Version 1

Abstract

Figures

1 Introduction

2 Materials and methods

2.1 Data collection

2.2 Identification of correlation, differential expressed, and prognosis related genes

2.3 Identification of HCC subtypes

2.4 Construction and validation of a lactylation-m6A related prognostic signature

2.5 Enrichment analysis

2.6 Immune cell infiltration, tumor mutation burden and tumor microenvironment analysis

2.7 The prediction of drug sensitivity

2.8 Cell culture and transfection

2.9 Wound healing and Transwell® assays

2.10 Statistical Analysis

3 Results

3.1 Analysis of lactylation-m6A related genes and DEGs in HCC patients

3.2 Identification of HCC subtypes based on prognostic IGs

3.3 Constructing a lactylation and m6A related prognostic model

3.4 The correlation of TMB and prognostic signature

3.5 Correlation analysis between the risk score and tumor microenvironment

3.6 Prediction drug sensitivity

3.7 The protein expression of risk genes

3.8 TCOF1- and HDAC1-silencing inhibited the cell migration and invasion

4 Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1