The significance of epigenetic regulation in the progression of development and disease is widely recognized. The introduction of single-cell sequencing technologies has facilitated the development of various methods for analyzing epigenetic regulation at different levels, including chromatin accessibility, DNA methylation, histone modifications, and nucleosome positioning, among others. These technologies have enabled researchers to investigate various aspects of epigenetics at the single-cell level [20, 51]. Such techniques have emerged as powerful tools for uncovering the distinctive epigenomic characteristics of rare cellular subtypes and the epigenetic diversity present within cellular populations.

Nucleosomes serve as the fundamental structural units of eukaryotic chromatin which consist of DNA wrapped around a core of histone proteins. The chromatin structure is further organized into a three-dimensional arrangement, which includes densely packed regions and more open, accessible regions. These distinct regions of chromatin structure play a crucial role in regulating gene expression. During replication and transcription processes, the tightly packed chromatin structure needs to be opened up to expose specific DNA sequences for regulatory factors to bind and carry out their functions. This opening up of the chromatin structure to allow regulatory factor binding is referred to as chromosome accessibility [52]. It involves the dynamic modulation of the chromatin structure, allowing access to the DNA. Studies have demonstrated that the accessibility of DNA sequences within the chromatin structure can influence gene transcription activity. Changes in chromatin accessibility can either promote or repress gene expression [52, 53]. This process is mediated by various mechanisms, including histone modifications, chromatin remodeling complexes, and interactions with transcription factors.

Chromatin accessibility is a field of active research and understanding its impact on gene expression has important implications in various cellular processes and diseases. By studying the accessibility of DNA within chromatin, scientists can gain insights into the mechanisms underlying gene regulation and potentially discover new targets for therapeutic interventions. The field of single-cell chromatin accessibility sequencing has made significant progress, enabling the detailed exploration of chromatin accessibility in individual cells shown in Fig. 4. Among the techniques developed for this purpose, single-cell ATAC-seq (scATAC-seq) has emerged as a powerful method. scATAC-seq leverages the Tn5 transposase’s sensitivity to identify open and accessible chromatin regions [54]. It allows for the investigation of chromatin accessibility landscapes at a single-cell resolution, providing insights into the epigenetic regulation and heterogeneity within tumors.

Insights into the dynamic regulation of chromatin accessibility and methods for single-cell chromatin accessibility sequencing

In addition to scATAC-seq, other techniques have been employed to assess chromatin accessibility at the single-cell level. For example, single-cell DNase-seq (scDNase-seq) utilizes DNase I digestion to identify accessible DNA regions in individual cells [55]. This method enables the detection of regions of open chromatin and provides valuable information about transcription factor binding and regulatory elements. Other new technologies are shown in Table 2.

These single-cell chromatin accessibility methods have been widely applied in tumor research, offering insights into the heterogeneity and regulatory dynamics within tumor cell populations. For instance, a study utilized single-cell ATAC-seq technology to analyze the cellular composition and state changes that occur during the transformation from healthy colon to precancerous adenomas to colorectal cancer (CRC). It revealed the cellular composition and state changes that occur during the transformation from healthy colon to adenomas and subsequently to CRC. In the cancerous state, the study observed T cell exhaustion, RUNX1-regulated cancer-associated fibroblasts, and increased accessibility associated with HNF4A motifs in epithelial cells. Furthermore, in sporadic CRC, the DNA methylation changes were strongly anti-correlated with the accessibility changes along this continuum, thus providing additional regulatory markers for molecular staging of polyps [55].

The process of tumor metastasis is a prominent contributor to mortality among individuals diagnosed with cancer [62, 63], and the utilization of scATAC-seq has significantly enhanced our comprehension of the underlying mechanisms involved in tumor metastasis. A study identified novel cell subpopulations with abnormally high CXCL14 expression levels in patients with breast cancer PL by transposase accessible chromatin (ATAC) sequencing (scATAC-seq) of breast cancer negative (NL) and positive lymph nodes (PL), and also identified potential regulators that may be associated with breast cancer lymph node metastasis, improving our understanding of the mechanism of lymph node metastasis of lymph node metastasis and provide a new prognostic marker for breast cancer lymphatic metastasis [64]. Another study confirmed that TCF7 promotes epithelial-mesenchymal transition (EMT) and that activation of EMT is a key process in cancer metastasis by performing single-cell RNA sequencing and scATAC-seq on tumors from patients with low risk of recurrence, high risk of recurrence, and recurrent bladder cancer [65].

The tumor microenvironment encompasses diverse cellular and structural constituents that exert a pivotal influence on tumor advancement and resistance to therapeutic interventions. Moreover, the examination of single-cell chromatin accessibility can be extended to the analysis of the tumor microenvironment. In one study, by performing scRNA-seq and scATAC-seq on ccRCC primary tumor tissues, investigators identified key regulatory molecules in the tumor microenvironment that mediate tumor progression and manipulate immune cell function, and further experimentally validated their role in tumor growth [66]. Another study identified regulatory mechanisms associated with CD8 T-cell depletion by analyzing scATAC-seq profiles of serial tumor biopsies before and after programmed cell death protein 1 blockade [67].Twenty-two human IDH mutant gliomas were analyzed by scATAC-seq, which explained how different subtypes of IDH mutant gliomas maintain different phenotypes and tumor microenvironments despite being derived from a common spectral hierarchy, and found that ATRX regulates glial identity and tumor microenvironment in IDH mutant gliomas [68].

scATAC-seq proves to be a valuable tool for investigating alterations in cellular response to therapeutic interventions throughout the course of tumor treatment. To improve the clinical outcome of CAR-T cell therapy, a study by scATAC-seq of sorted T-cell subsets from seven patient, it was found that IRF7-regulated chronic IFN signaling was associated with poor persistence of CAR-T cells in T-cell subsets and that TCF7 regulators were relevant not only in maintaining naive and early memory T-cell states, but also in maintaining a good phenotype in effector cell lineages play a role [69]. Another study, by scATAC-seq of 12 breast cancer patients, found that the transcription factor GRHL2 cooperates with FOXA1 to initiate endocrine resistance and that epigenetic heterogeneity may contribute to endocrine resistance in breast cancer patients [70].

Table 2 presents the unique advantages and limitations of single-cell chromatin accessibility in terms of characteristics, throughput, and applicability.

DNA methylation is a modification in which –CH3 methylation modification occurs on cytosine bases [71], specifically 5mC (5-methylcytosine) in this article. This methylation process is catalyzed by the DNA methyltransferase (DNMT) family, which adds methyl to the cytosine of the 5′ terminal CpG dinucleotide of human genes. A CpG island is a region in the 5′ → 3′ direction of a DNA sequence that is rich in CG dinucleotides. CpG islands can be found in the promoters of more than two-thirds of all genes and in most cases these CpG islands are in an unmethylated state. CpG islands act as definitive markers of DNA methylation and act as key switches in epigenetic regulation [72], restricting gene expression in the presence of methylation. Studies have shown that methylation of CpG islands plays an important role in transcriptional regulation and is usually altered during malignant transformation. Approximately 5–10% of normal unmethylated CpG island promoters are aberrantly methylated in various cancer genomes. Methylation of promoter CpG islands mediates the phenomenon of gene silencing observed during tumorigenesis [73, 74]. In addition, dysregulated DNA methylation is considered a hallmark of cancer. Genomic demethylation and gene-specific hypermethylation are prevalent in oncogenes and tumor suppressor genes [75,76,77] (Fig. 5).

Insights into the dynamic regulation of DNA methylation and methods for single-cell DNA methylation sequencing

With the advancement of next-generation sequencing (NGS) technologies, high-throughput methylation sequencing methods have made significant progress, enhancing the accessibility and efficiency of sequencing. Among these methods, bisulfite sequencing is considered the “gold standard” for DNA methylation analysis due to its high accuracy and single-base resolution [78, 79]. This technology includes whole-genome bisulfite sequencing (WGBS), which evaluates the extent of methylation within CpG islands by converting unmethylated cytosine © to uracil (U) while leaving 5-methylcytosine (5mC) unchanged [80]. WGBS provides in-depth understanding of the DNA methylation patterns across the entire genome and has revolutionized our understanding of DNA methylation. To reduce costs and increase sample throughput, researchers have developed methods that target specific regions for methylation sequencing, such as reduced representation bisulfite sequencing (RRBS) [81]. RRBS utilizes restriction enzymes to enzymatically cleave the genome, reducing its complexity during sequencing and enriching the analysis for important regulatory regions like promoters and CpG islands where detailed methylation analysis can be performed.

However, previous methods relied on bulk sequencing, which averaged the methylation information of cell populations, unable to resolve the heterogeneity present within individual cells. With the development of single-cell sequencing technologies, single-cell methylation sequencing has become feasible. The first method based on single-cell bisulfite sequencing was single-cell methylation genome sequencing (scRRBS), which employs enzymatic cleavage to generate CpG-rich DNA fragments for subsequent library construction and sequencing [82]. However, the harsh conditions used in bisulfite conversion can lead to DNA degradation, resulting in DNA loss and reduced sequencing quality, ultimately affecting the data yield. To address this issue, researchers have developed PBAT (post-bisulfite adaptor tagging) to mitigate the loss caused by degradation. Other single-cell methylation sequencing methods have also been developed, as shown in Table 3.

Methylation events have a major impact on the regulation of cell fate, and single-cell DNA methylation sequencing has provided key new insights into the important issue of tumor heterogeneity. For instance, a study utilized single-cell bisulfite sequencing (scBS-seq) technology to characterize partial methylation domains (PMDs) within individual cells of colorectal cancer. The results revealed that over half of the genome was covered by PMDs, and Gain-PMDs, a specific subtype, exhibited a higher coverage of protein-coding genes. Furthermore, the study unveiled substantial epigenetic heterogeneity among different cells within the same tumor and demonstrated that DNA methylation in cells is influenced by the tumor microenvironment [97].

Another study emphasizes the significance of genetic and epigenetic heterogeneity within tumors and its impact on the evolutionary trajectory of cancer. The researchers utilized single-cell bisulfite sequencing analysis (MscRRBS) to investigate this heterogeneity in chronic lymphocytic leukemia (CLL). Their findings revealed that CLL cells exhibit high rates of epigenetic mutations, while showing minimal variation in mutation rates among individual cells. Through comprehensive single-cell analyses, the study elucidated the lineage diversity of CLL cells and their evolutionary patterns following treatment. Notably, the researchers observed specific lineage biases during therapy. By integrating genetic, epigenetic, and transcriptional information at the single-cell level, this study reconstructed the genealogical history of CLL, thereby providing valuable insights into the understanding of tumor development [98].

There are also articles that investigated the DNA methylation profiles of circulating tumor cells (CTCs) using the scBS-seq technique and revealed the subclonal structure, evolutionary history and classification of tissue-specific DNA methylation profiles in CTCs. The results indicate the heterogeneity of DNA methylation in CTCs and reveal the epigenetic regulatory mechanisms in cancer metastasis [99].

Prior epidemiological studies have established a significant association between the consumption of food contaminated with aflatoxin B1 and the incidence of hepatocellular carcinoma. Scientists utilized single-cell RRBS technology to investigate the hepatotoxic mechanism induced by aflatoxin B1 (AFB1) in S phase-arrested L02 cells. The study found that AFB1 caused apoptosis and S phase arrest in L02 cells, reduced mitochondrial membrane potential, increased reactive oxygen species generation, and led to an increase in DNA methylation levels. Through single-cell RRBS analysis, it was revealed that DNA methylation, regulated by the gonadotropin-releasing hormone receptor pathway, Wnt signaling pathway, and TGF-beta signaling pathway, was involved in the hepatotoxic mechanism induced by AFB1 in S phase-arrested L02 cells [1].

Hannah Demond et al. utilized single-cell bisulfite sequencing (scBS-seq) to investigate DNA methylation abnormalities caused by maternal effect mutations in the subcortical maternal complex (SCMC) of human oocytes. These mutations are associated with early embryonic failure, gestational abnormalities, and recurrent pregnancy loss. The researchers observed a genome-wide deficiency of DNA methylation in the oocytes of patients with SCMC mutations compared to normal oocytes. Both the germline differentially methylated regions (gDMRs) of imprinted genes and other sequence features that are normally methylated in oocytes were affected, indicating a lack of selectivity toward imprinted genes. The degree of methylation loss varied across different genomic features. Furthermore, analysis of a preimplantation embryo and molar tissue from the same patient revealed persistent methylation defects at imprinted genes after fertilization, while non-imprinted regions of the genome showed near-normal methylation levels after implantation. These findings emphasize the critical role of the SCMC in de novo methylation in the female germline and provide valuable insights into imprinting defects and potential therapeutic strategies [100].

Nevertheless, single-cell methylation sequencing does have limitations. Firstly, due to the limited amount of DNA within a single cell, single-cell sequencing often faces challenges of low coverage and sparse results. Secondly, compared to traditional bulk cell sequencing, single-cell sequencing may exhibit higher technological noise. For example, the DNA amplification process can introduce biases, resulting in oversampling or undersampling of certain methylation sites and impacting the final analysis results. Additionally, during the single-cell collection process, there may be loss of cellular information, such as spatial relationships between cells, and some cells may not be successfully sequenced due to isolation process-related damage. In conclusion, single-cell methylation sequencing is crucial in uncovering the methylation heterogeneity within individual cells. However, challenges and opportunities for improvement remain in terms of throughput, sequencing depth, and accuracy. As technology continues to advance and improve, we can expect wider applications of single-cell methylation sequencing in life science research.

Table 3 presents the unique advantages and limitations of single-cell DNA methylation in terms of characteristics, throughput, and applicability.

Within a cell, genomic DNA is not just a string of linear sequences but exhibits a highly complex three-dimensional (3D) spatial structure. This structure is largely dependent on histones, octamers composed of two units each of four core components (H2A, H2B, H3, H4) [101]. These histones possess protruding “tail” structures, which can be regulated via a series of chemical modifications.

Histone modifications represent a key epigenetic mechanism, encompassing, but not limited to, acetylation, methylation, phosphorylation, glycosylation, ubiquitination, and nitrosylation. These modifications are usually catalyzed by specific enzymes, such as histone methyltransferase (HMT) and histone acetyltransferase (HAT). These enzymes add specific chemical groups to the amino acid residues of the histones or remove them, thus altering the state of the chromatin [102].

HMT and HAT can be perceived as “writers,” responsible for adding chemical marks. In turn, histone demethylase (HDM) and histone deacetylase (HDAC) can be seen as “erasers,” tasked with erasing these marks [103]. This series of modifications affects the compactness of the chromatin structure, thereby changing its interaction with DNA, and ultimately leading to the activation or silencing of genes.

The advent of next-generation sequencing (NGS) technologies has been a significant leap forward in understanding gene regulation and epigenetic mechanisms. Early techniques such as chromatin immunoprecipitation (ChIP) microarrays paved the way for more advanced methods, most notably Chromatin Immunoprecipitation Sequencing (ChIP-Seq). Compared to its predecessors, ChIP-Seq offers unparalleled resolution down to the single base-pair level, reduced methodological artifacts, and comprehensive genomic coverage [104]. ChIP-Seq is the gold standard for genome-wide analyses of DNA–protein interactions, histone modifications, and nucleosome positioning. While it offers a wealth of data, traditional or ‘bulk’ ChIP-Seq falls short in assessing the variability in chromatin states across individual cells. This limitation was addressed in 2015, when a landmark paper by David Weitz and Bradley E Bernstein in Nature Biotechnology introduced the concept of single-cell ChIP-Seq108. This innovation permitted the analysis of histone modifications at the single-cell level, thereby unveiling a new dimension of chromatin state heterogeneity. Subsequent to this pioneering work, various methods have emerged to probe chromatin states at single-cell resolution. Techniques such as scCUT&Tag [105], CoBATCH [106], and scChIC-seq [107] have further refined our understanding of chromatin dynamics. Each of these methods summarized in Table 4 and Fig. 6.

Insights into the dynamic regulation of Histone Modification and methods for single-cell Histone Modification sequencing

The significance of single-cell protein modifications lies in revealing the identity and differentiation state of cells, unraveling cellular heterogeneity, studying disease mechanisms, and guiding treatment response and drug discovery. Analyzing protein modification patterns in individual cells helps to understand cellular function and regulation, identify cell-to-cell differences, and provide insights into disease mechanisms for potential therapeutic targets.

A study utilized high-throughput single-cell ChIP-seq technology to investigate chromatin heterogeneity and drug resistance in breast cancer. The researchers employed a microdroplet microfluidic platform to sequence and analyze the chromatin states of thousands of individual cells at single-cell resolution. The study revealed that in untreated drug-resistant tumors, there exists a subset of cells that share chromatin markers with drug-resistant cells, and these cells have lost the chromatin marker H3K27me3 associated with genes promoting drug resistance. This single-cell ChIP-seq technology offers a novel approach to studying the role of chromatin heterogeneity in cancer and other diseases, and aids in uncovering the regulatory mechanisms involved in cellular differentiation and development [116].

There are also study through scCUT&Tag method, combined with scalable nanopore and droplet-based single-cell platforms, was employed to analyze specific chromatin regions in individual cells. The focus was on analyzing the polycomb group (PcG) silencing regions marked by the histone modification H3K27me3 as an orthogonal approach to identify cell states based on chromatin accessibility. Results showed that scCUT&Tag analysis of H3K27me3 could distinguish different cell types in human blood and generate cell type-specific PcG landscapes in heterogeneous tissues. Furthermore, the study utilized scCUT&Tag to analyze H3K27me3 in brain tumor patients before and after treatment, identifying cell types in the tumor microenvironment and revealing heterogeneity in PcG activity between primary samples and post-treatment [117].

Another study utilized automated CUT&Tag chromatin profiling to investigate the impact of KMT2A oncofusion proteins in leukemias. By mapping fusion-specific targets across the genome, the researchers identified common and tumor-subtype-specific sites of aberrant chromatin regulation. They found that certain binding sites for KMT2A oncofusion proteins exhibited cell-to-cell heterogeneity and were associated with lineage plasticity. Additionally, they discovered that abnormal enrichment of H3K4me3 in gene bodies could be targeted by Menin inhibitors. The integration of automated and single-cell CUT&Tag techniques enabled the identification of epigenomic heterogeneity within patient samples and the prediction of therapeutic sensitivity [114].

Over the years, we have gradually come to recognize the complexity and importance of histone modifications and chromatin states in gene regulation. The emergence of single-cell technologies is a significant breakthrough, allowing us to explore previously uncharted layers of epigenetic regulation. New technologies are expected to emerge that integrate the advantages of existing methods, providing higher resolution and throughput, and possibly reducing costs as well. Furthermore, the continued development of computational analysis and machine learning will help in parsing increasingly complex datasets. In short, the field of single-cell histone modification is still evolving and holds the promise of providing us with a deeper understanding of gene regulation. It may also inspire our insights into cellular processes and disease mechanisms.

Table 4 presents the unique advantages and limitations of Histone Modification sequencing in terms of characteristics, throughput, and applicability.

Nucleosome positioning refers to the precise localization of nucleosomes, which are structural units composed of an octamer of histone proteins and the wrapped DNA, on the genome [109]. In eukaryotic chromosomes, the binding of DNA to histones is not static, and the accurate determination of nucleosome positions on the genome, known as nucleosome positioning, is crucial for maintaining genome structure and function.

The regulation of nucleosome positioning is closely associated with the spatial organization of chromatin, DNA replication, transcription, and gene expression regulation [118]. Nucleosome positioning can influence the chromatin state and accessibility, thereby impacting the transcriptional activity of genes. The positions of nucleosomes on the genome can affect the binding of transcription factors and the accessibility of promoters, determining the transcriptional levels and patterns of genes.

The study of nucleosome positioning can be conducted using various experimental techniques and computational methods. Among them, micrococcal nuclease sequencing (MNase-seq) is a widely used approach that involves the digestion of chromatin with micrococcal nuclease to degrade nucleosome structures, releasing the core DNA of nucleosomes, which can then be sequenced using high-throughput sequencing technology [56]. By analyzing the sequencing data, the positions and positioning patterns of nucleosomes can be determined.

Recent studies have demonstrated that even within a homogeneous population of cells, different cells exhibit significant heterogeneity in chromatin states [119, 120], which may be related to heterogeneity in chromatin accessibility. scMNase-seq is a high-throughput sequencing technique used for studying nucleosome positioning at the single-cell level. It can uncover the heterogeneity of nucleosomes between cells and investigate cell-specific chromatin states and gene regulatory mechanisms [56].

In one research, through single-cell MNase-seq analysis, researchers identified distinct accessible chromatin regions in all lymphoid progenitor cells (ALP), early ILC progenitors (EILP), and ILC progenitors (ILCP). Within EILP, different subpopulations were identified, indicating their potential to differentiate into either dendritic cell lineage or ILC lineage based on epigenetic profiles. The researchers found that the transcription factors TCF-1 and GATA3 co-bound with lineage defining sites (LDS-Is) associated with ILC, while PU.1 binding was enriched in LDS (LDS-As) associated with alternative dendritic cell fate. TCF-1 and GATA3 were found to be essential for the epigenetic priming of LDS at the EILP stage. These findings suggest that the fate of multipotent progenitors during the differentiation into ILC and hematopoietic stem cells is pre-defined by their epigenetic state. The presence of distinct subpopulations within multipotent progenitors and their regulation by key transcription factors highlight the heterogeneity of cells that contribute to lineage specification. The application of single-cell MNase-seq technology enables a deeper understanding of the epigenetic changes during cellular differentiation, shedding light on the molecular mechanisms and role of transcription factors in determining cell fate [121].