Unveiling aging dynamics in the hematopoietic system insights from single-cell technologies

Abstract

As the demographic structure shifts towards an aging society, strategies aimed at slowing down or reversing the aging process become increasingly essential. Aging is a major predisposing factor for many chronic diseases in humans. The hematopoietic system, comprising blood cells and their associated bone marrow microenvironment, intricately participates in hematopoiesis, coagulation, immune regulation and other physiological phenomena. The aging process triggers various alterations within the hematopoietic system, serving as a spectrum of risk factors for hematopoietic disorders, including clonal hematopoiesis, immune senescence, myeloproliferative neoplasms and leukemia. The emerging single-cell technologies provide novel insights into age-related changes in the hematopoietic system. In this review, we summarize recent studies dissecting hematopoietic system aging using single-cell technologies. We discuss cellular changes occurring during aging in the hematopoietic system at the levels of the genomics, transcriptomics, epigenomics, proteomics, metabolomics and spatial multi-omics. Finally, we contemplate the future prospects of single-cell technologies, emphasizing the impact they may bring to the field of hematopoietic system aging research.

INTRODUCTION

Aging constitutes a major susceptibility factor for numerous chronic maladies in humans, including hematologic disorders, cardiovascular ailments, cancer and neurodegenerative conditions. The aging process entails substantial alterations in cellular constitution, organ dimensions and functionality across diverse tissues and organs within an organism. Notably, the hematopoietic system, comprising blood cells and the associated bone marrow niche, intricately participates in hematopoiesis, coagulation, immune modulation and other physiological phenomena [1, 2]. Hematopoietic stem cells (HSCs) exhibit the remarkable capacity for both self-renewal and differentiation, ensuring the perpetual production of the entire spectrum of blood cells throughout their life cycle. The aging process instigates manifold transformations within the hematopoietic system. Extant research delineates that age-associated changes in the bone marrow microenvironment and HSCs engender disruptions in tissue homeostasis, manifesting in discernible disparities between youthful and aged cohorts. These disparities encompass shifts in total cellular quantity, lineage-specific differentiations, cellular constitution and functional attributes of HSCs [3–6]. These are a series of risk factors leading to hematopoietic diseases, including clonal hematopoiesis, immune aging, myeloproliferative neoplasms and leukemia [7–9]. Consequently, unraveling the intricate molecular mechanisms underlying the aging of HSCs assumes paramount importance. This endeavor holds the key to identifying potential intervention targets aimed at prolonging hematopoietic healthspan. Moreover, it bears relevance in the therapeutic landscape for addressing immune system deficiencies, age-related inflammatory pathologies and hematologic maladies.

In the realm of experimental interpretations, both in vitro and in vivo, a prevailing assumption persist, namely the uniformity of utilized cells or tissues. However, such a conventional approach often neglects inherent cellular variances, resulting in the inadvertent loss of crucial information. The complex multitude of various cell types in hematopoietic system, from HSCs to mature cells, within both normal physiological processes and pathological states, makes the research on it particularly challenging. In the preceding century, pioneers in the scientific community initiated the sequencing of the entire transcriptome at the single-cell level [10]. In 2009, Tang laid the foundation for the single-cell mRNA-Seq method [11]. Over the ensuing decade, as this technology matured and became commercially accessible, it underwent rapid evolution. What was once confined to specialized research laboratories has now proliferated into a diverse array of commercial applications. While single-cell transcriptomics stands as an exemplary tool for quantifying gene expression variations at the single-cell tier, it has become increasingly evident that it falls short of fully addressing the diverse requirements of researchers across multifarious domains. Over this decade-long trajectory, single-cell technology has not merely transcended the bounds of transcriptomics; it has extended its purview into genomics, metabolomics, epigenomics and beyond [12–15]. Single-cell technology has been used in many fields, such as analyzing steady-state cardiovascular cell diversity and cardiovascular disease targets [16], explaining the molecular mechanisms and spatial characteristics of brain aging [17, 18] and revealing the mechanisms of cancer [19–22].

Single-cell technology, with its unparalleled capacity to unravel intricate cellular compositions, serves as a potent instrument for dissecting the aging dynamics of the hematopoietic system, thus advancing our comprehension of the underlying mechanisms of aging. This review is focalized on recent revelations made possible by single-cell technologies in the investigation of aging dynamics spanning diverse lineages within the hematopoietic system across multiple species. We present a brief overview of each single-cell technique and its applications in studying hematopoietic aging (Table 1) and a timeline highlighting key findings from single-cell studies of the hematopoietic system (Fig. 1).

GENOMIC ALTERATIONS CORRELATED WITH HEMATOPOIETIC AGING

The accrual of DNA damage is commonly acknowledged as a characteristic feature of the aging process [41]. With advancing age, there is a gradual accrual of somatic mutations across tissues. While the majority of these mutations may have minimal functional impact, some mutations can confer new phenotypes on cells, such as an increased rate of division. In the hematopoietic system, specific functional changes can occur when certain HSCs acquire mutations that enhance their proliferation ability, leading to the development of a substantial number of mature blood cells—a phenomenon known as clonal hematopoiesis [42, 43]. Research indicates that clonal hematopoiesis is associated with aging: Somatic clones are detectable in a small proportion of healthy individuals under the age of 40, but the likelihood of their occurrence rises with age [44]. Clonal hematopoiesis is also linked to an elevated risk of hematological malignancies [45, 46]. Hematologic malignancies exhibit cellular heterogeneity, with rare cell populations, including drug-resistant cell clones and leukemia stem cells, playing a crucial role in disease progression. Given this complexity, single-cell genomics approaches are particularly suitable for studying these malignancies [47].

Genomic DNA

Single-cell techniques are currently employed to investigate the association between somatic mutations and age-related diseases. Unlike single-cell RNA sequencing (scRNA-seq), the advancement of single-cell DNA sequencing (scDNA-seq) methods poses a greater challenge due to the inherent complexity of dealing with only two copies of genomic DNA in a single cell. In 2012, Zong [48] pioneered the development of multiple unified whole-genome amplification (WGA) methods, including multiple annealing and loop-based amplification cycles (MALBAC), enabling single-cell whole-genome sequencing (scWGS). Moreover, several methods have emerged to simultaneously perform scRNA-seq and scDNA-seq on individual cells. Instances of such methods include DR-Seq and G&T-seq, both relying on comprehensive cell lysis followed by genome isolation [49, 50]. Additionally, techniques such as isolation of polyadenylated RNA from DNA and direct nuclear labeling and RNA-seq have been employed in this context [51, 52].

TARGET-seq facilitates simultaneous mutation and RNA sequencing analysis in single cells. It uses flow cytometry combined with cell surface proteomics to correlate gene expression and mutational status within specialized cell populations [39]. An examination of non-mutated cellular HSCs has yielded valuable insights into the disruption of normal hematopoiesis in myeloproliferative neoplasms (MPNs), revealing that cells outside the MPN clone also exhibit aberrant gene expression [23]. Genotyping of Transcriptomes (GoT) is presently employed to distinguish mutation-positive cells from mutation-negative cells of the same type. The simultaneous analysis of mutation status and transcriptome using GoT was initially applied to hematopoietic stem and progenitor cells (HSPCs) from CALR-mutated essential thrombocythemia (ET) patients [24]. A study unveils intertumor heterogeneity by conducting single-cell copy number analysis and genotyping of HSPCs isolated from the bone marrow of patients with myelodysplastic syndromes (MDS) [25]. Another study employed targeted single-cell sequencing of whole myeloid cells and multipotent progenitor cells to unveil the genetic foundation of disease initiation in T-cell acute lymphoblastic leukemia (T-ALL), and found that mutation in T-ALL may originate from multipotent progenitor cells [53]. In another study, single-cell genome sequencing was employed to delineate clonal evolution in T-ALL patients during diagnosis and treatment, and revealed that minor clones evolved into clinically relevant major clones later in the disease course [54]. Utilizing single-cell genome and transcriptome analyses, a study proposes that mutations in LCP1 and WNK1 could represent novel mechanisms propelling the development of chronic lymphocytic leukemia (CLL) [26]. Gawad sequenced individual tumor cells from patients with ALL and detected co-dominant clones in the majority of cases. This sequencing approach offers insights into the disease at a single-cell resolution, paving the way for potential novel treatment strategies [55].

Mitochondrial DNA

Each cell harbors hundreds to thousands of mitochondrial DNA (mtDNA) copies, which replicate independently of nuclear genomic DNA [56]. In comparison to nuclear DNA, mtDNA accumulates mutations at a rate 5–10 times higher, primarily attributed to the inability to undergo DNA repair and frequent exposure to reactive oxygen species [57], contrary to the conventional belief that individuals predominantly possess identical mitochondrial genomes. Currently, research is underway to explore the origins of human mtDNA variations and their association with human diseases, encompassing cancer, neurodegenerative diseases and aging [58–62]. Notably, hundreds of mtDNA mutations have been identified as causes of various human diseases [63]. However, the contribution of mtDNA mutations to these processes remains unclear, and studying mtDNA mutations and their associated pathogenic effects at the single-cell level is essential [64]. Several methods have been developed to detect mtDNA mutations at the single-cell level [27, 65–69]. One study performed a systematic evaluation of mtDNA mutations in B lymphocytes and monocytes of an elderly individual at the single-cell level, and found that over 50% of cells carried mtDNA mutations, and over 30% those mutations were predicted to be highly pathogenic, suggesting that mtDNA mutations may constitute a significant source of mitochondrial function decline with age [27]. Studies demonstrated higher mtDNA biosynthesis in acute myeloid leukemia (AML) blast cells compared to normal hematopoietic cells [70]. Resistance to imatinib in chronic myelogenous leukemia (CML) is linked to heightened instability in mtDNA and elevated levels of reactive oxygen species [71]. The integrity of mitochondria is crucial to HSC function, and mutations in mtDNA may lead to diseases. While clear mitochondrial abnormalities are observed in leukemia cells, their role in leukemia development and progression remains relatively unknown. Single-cell sequencing analysis enables the tracking of specific HSC clones and reveals mutation patterns in hematopoietic cells at the single-cell level, which will provide new insights into the occurrence of mtDNA mutations in HSCs and enhance our understanding of their role in aging and disease.

TRANSCRIPTOMIC ALTERATIONS CORRELATED WITH HEMATOPOIETIC AGING

Comparing the transcriptome of cells with disparate ages within the same cell type is a common strategy to investigate the molecular mechanisms underlying aging. Given the hematopoietic system’s diversity in cell populations, leveraging single-cell transcriptomics provides a necessary and valuable approach. Single-cell level analysis can determine whether age-related changes occur in specific cell type or homogeneously present across all cell populations.

Age-related gene expression changes in HSCs

Before the widespread adoption of single-cell sequencing technology, considerable attention was devoted to investigating alterations in HSC function during aging, yet the underlying molecular mechanisms remained incompletely understood. Presently, numerous studies leverage single-cell sequencing to scrutinize the transcriptome of HSCs across various age groups, providing more precise insights. This approach aids in unraveling age-specific changes in HSC gene expression, exposing factors like lineage bias, imbalances in self-renewal and differentiation, and the molecular mechanisms underpinning phenotypes.

Transcriptomic analysis at the single-cell level offers distinct advantages in elucidating the phenotypic changes in HSCs during aging. Researchers frequently employ single-cell techniques to investigate HSC function in mouse models. Single-cell RNA-seq (scRNA-seq) was used to analyze transcriptome changes in HSPCs in young and old mice. The findings revealed that the variability within each cell population was predominantly influenced by the cell cycle. Notably, the number of cells in the G1 phase was lower in old long-term hematopoietic stem cells (LTHSCs) compared to young LTHSCs, suggesting an accelerated progression through the G1 phase, which is consistent with models proposing that changes in cell cycle progression in relation to HSC self-renewal and differentiation are associated with aging [28, 29]. In another study, a thorough analysis of the single-cell transcriptomes and function of mouse HSCs revealed that aging correlates with an upregulation of platelet-specific genes in HSCs. [30]. Hérault conducted a comparative analysis of HSPCs in young and old mice, revealing alterations in the expression of genes associated with loss of differentiation and heightened hemostatic characteristics during aging. Notable changes included the upregulation of Nupr1, Vwf and Clu and undifferentiated HSPC markers, along with the downregulation of HSC differentiation genes. These findings align with previously reported shifts in the differentiation potential and platelet bias of aged HSPCs [31]. One study compiled 16 published and unpublished datasets, determining that dysregulated senescence genes predominantly consist of membrane-associated transcripts, including numerous cell surface molecules not previously linked to HSC biology. Selp is not only associated with reduced HSC function but also serves as a marker of aging HSCs. Furthermore, single-cell transcriptome analysis revealed the presence of a specialized population of HSCs in aging bone marrow with transcriptional profiles similar to those of young HSCs [72]. In a previous study, Kirschner characterized the heterogeneous aging of HSCs through single-cell transcriptome sequencing analysis, identifying a subset of aged HSCs exhibiting signs of functional failure. The expansion of the aged HSC subpopulation through increased proliferation was linked to decreased HSC function, underscoring the connection between prolonged proliferation and diminished HSC functionality [73]. Studying gene expression at the isoform level is required to fully understand transcriptional cellular heterogeneity. ScRNA-seq can annotate cells and cell types with isotype-level expression information, allowing a comprehensive understanding of the complexity of cellular transcriptional and functional phenotypes. In a recent study, Mincarelli et al. presented an integrated short- and long-read single-cell RNA-seq analysis of haematopoietic stem and progenitor cells. They demonstrated that over half of detected genes are expressed as multiple, often functionally distinct, isoforms, including many transcription factors and key cytokine receptors. Such an approach not only facilitates the integration of previously unlabeled isoforms into reference transcriptomes but also enhances the accuracy of annotating cell type-specific isoform expression [32].

In addition to single-cell analysis in mouse models, researchers also directed their attention to human bone marrow cells. The data revealed variations among healthy donors and demonstrated age-related changes in cell population frequencies [74]. ScRNA-seq is very conducive to the analysis of HSC heterogeneity, but provide limited temporal information. HSPC biology is highly dynamic, which urgently necessitates the high-resolution analysis of the quantity, differentiation, proliferation and other dynamic processes of HSCs at different time points in vivo. In a recent study, a real-time and quantitative murine bone marrow hematopoietic cell flow model was constructed based on time-series scRNA-seq and continuous labeling techniques. This model was employed to reveal various self-renewal processes and differentiation characteristics within different lineages, bridging molecular processes with cellular behavior [75]. The innovative aspect of this model lies in its connection of high-resolution molecular information with tissue-scale cellular behavior, filling a void in the lack of in vivo dynamic models in the hematopoietic system and holding potential applications in biological fields beyond hematopoiesis.

Gene expression changes in mature blood cells and age-related diseases

Age-related changes in the mature transcriptome can potentially elucidate the functional alterations observed during human aging. The Tabula Muris Consortium employed scRNA-seq to construct a comprehensive map of age-related changes in the organs and tissues of mice across different age groups. This represents one of the most extensive analyses to date of the mammalian aging process at single-cell resolution. Results showed that white blood cells from aged mice showed increased expression of pro-inflammatory markers, while decreased expression of anti-inflammatory markers [76]. Studies have also showed a 2-fold reduction in the fraction of naive CD8 T cells in peripheral blood mononuclear cells in older adults compared to younger adults [77, 78]. In a study by Hashimoto, single-cell transcriptome analysis of peripheral blood mononuclear cells from centenarians and young controls revealed that a unique subset of cytotoxic CD4 T cells was significantly increased in centenarians [79]. This subset may contribute to longevity by preserving immune responses against infection and disease. Additionally, another study reported a progressive increase in B cell clonality in aging mice [80]. The results of these single-cell studies help explore the antigenic specificity of aging-related cell subsets and their inflammatory functions in aging organisms.

The elderly population faces an elevated risk of malignant hematopoiesis [81]. Hematopoietic malignancies such as leukemia, lymphoma, multiple myeloma (MM), MPN and MDS are prevalent and pose a significant threat to patient health and survival. Employing scRNA-seq methods can contribute to a more profound understanding of these diseases and facilitate the discovery of more effective treatment strategies.

Alice and colleagues developed a method that combines highly sensitive mutation detection with whole-transcriptome analysis of the same single cells. They applied this technique to analyze single cells from CML patients, revealing the heterogeneity of CML-SCs. Furthermore, a crisis-specific stem cell population among blasts was identified, present in chronic-phase CML-SC subclones of patients who subsequently developed blast crisis [82]. Zhang conducted scRNA-seq analysis on individual cells isolated from the peripheral blood of CML patients at various stages of treatment, and found that imatinib treatment resulted in substantial alterations in leukocyte populations in both responders and non-responders, and gene expression changes in diverse cell populations, underscoring the potential utility of peripheral blood as a diagnostic tool for assessing tyrosine kinase inhibitor sensitivity in CML patients [83]. Ma integrated scRNA-seq profiles and network analysis of CML stem cells to unravel the mechanisms underlying different tyrosine kinase inhibitor responses. They found that a set of genes in various types of stem cells from CML patients was consistently differentially expressed compared to HSCs [34]. Vaidehi Krishnan identified pre-treatment signatures of primary imatinib resistance in CML through a single-cell atlas [84]. Wang discovered significant changes in multiple cellular functions of cells carrying SF3B1 mutations in CLL, affecting multiple cellular functions. They defined LCP1 and WNK1 mutations as new drivers of CLL development [85]. Van Galen analyzed AML heterogeneity using scRNA-seq, with a specific focus on FLT3 genotypes in malignant cell clusters [86]. Lohr et al. also used scRNA-seq to identify genes that are differentially expressed between circulating tumor cells from different MM patients, which can be used for the diagnosis and classification of MM [87]. By combing scRNA-seq and single-cell genotyping, Tong found that HSCs with JAK2 mutations showed a bias toward megakaryocyte differentiation, and this type of mutation was most common in MPN [88].

EPIGENETIC ALTERATIONS CORRELATED WITH AGING IN HEMATOPOIETIC SYSTEM

Epigenetics represents a reversible mechanism influencing gene expression which does not involve alterations of DNA sequence [89], and holds a pivotal role in numerous biological processes [90]. A study analyzed epigenomic and transcriptomic changes during normal human aging, and, with the help of scRNA-seq analysis, confirmed that the observed age-associated changes in the epigenetic program are unlikely to stem from the expansion of a pre-existing subpopulation within the bone marrow, but rather appear to result from true epigenetic reprogramming [33]. Epigenetic changes associated with aging encompass DNA methylation, chromatin remodeling and post-translational modifications of histones. Aging can induce alterations the epigenetic landscape within HSCs [91–93].

DNA modification

DNA methylation, the earliest studied epigenetic mark, plays a crucial role in various biological processes [94, 95]. Hui employed single-cell bisulfite sequencing (scBS-Seq) to explore the methylome of cells in the HSC compartment of both mice and humans, revealing the presence of distinct epigenetic subpopulations [38]. During the differentiation of HSCs into multipotent progenitors and cells of various lineages, DNA methylation undergoes extensive changes [96]. The aging process is associated with a widespread loss of methylation across the genome [97, 98]. In a study investigating the pathogenic mechanisms of aging in whole blood, MBD-based capture and sequencing technology identified 70 age-related differentially regulated methylation regions in 738 individuals [99].

The DNMT and TET protein families are crucial regulators of DNA methylation and demethylation processes, and their expression differs between young and old HSCs [100, 101]. Initial studies revealed that HSCs with reduced DNMT1 activity in mice could differentiate into myeloid erythrocytes but not lymphocytes, emphasizing the critical role of DNA methylation in HSC self-renewal activity [102]. Specific deletion of DNMT1 directly links DNA methylation to HSC senescence, resulting in a lineage bias toward myelopoiesis and self-renewal defects, characteristic of natural HSC senescence [103]. Wall found that as HSCs undergo differentiation into a lineage-committed state characterized by enhanced methylation of stemness genes and diminished methylation of lineage-specific genes. This suggests that aberrant activation or inactivation of demethylases or methylases may disrupt this balance, underscoring the crucial role of DNA methylases in HSC homeostasis [36, 104]. The TET demethylase is also implicated in aging. Quantitative mass spectrometry analysis revealed that 5hmC levels decrease during HSC aging in both mice and humans, indicating reduced 5hmC levels as an epigenetic mark of HSC aging [105, 106].

Histone modifications

Histone modifications play a crucial role in influencing the gene expression patterns of senescent HSCs [100]. Comprehensive genomic investigations have revealed numerous age-associated genes subject to regulation during the aging process. This includes lysine-specific demethylases, which exhibit a decline in expression levels with advancing age [107–110]. Comparative analyses between young and senescent HSCs reveal broader H3K4me3 peaks in senescent HSCs associated with HSC identity and self-renewal genes. Additionally, DNA methylation of transcription factor binding sites associated with differentiation is increased. The distribution of H3K27me3 undergoes significant changes, with numerous promoter regions displaying an increase in H3K27me3 marks [101]. Histone acetylation has also been implicated in HSC senescence. Under stress conditions, the deletion of the deacetylase SIRT1 in HSCs leads to an increase in HSC numbers, ultimately resulting in DNA damage and HSC depletion in mice [111]. Furthermore, young HSCs exhibit high levels of H4K16 acetylation, while a subset of senescent HSCs shows a significant decrease in H4K16Ac [112].

It is evident that alterations in histone modifications impact the state of HSCs, and researchers have utilized single-cell technology to delve into this relationship. In a recent study, Hui observed that BRD4 impairs HSC self-renewal and differentiation in the hematopoietic system. Transposase accessible chromatin sequencing (ATAC-seq) analysis reveals increased chromatin accessibility in Brd4^Δ/Δ HSC/HPC [113]. Cheung introduced an epigenetic landscape analysis using Time-Of-Flight cytometry, a single-cell method based on mass spectrometry cytometry to analyze various histones in single cells. This approach revealed a persistent increase in cell-to-cell variability in chromatin markers in human immune cells, a hallmark of immune cell senescence. The single-cell methodology offers novel insights into the role of epigenetic regulation in hematopoiesis, immune cell function and the aging of the immune system. Additionally, it reveals aberrant epigenetic patterns linked to immune-mediated diseases [114].

Chromatin remodeling

Single-cell assay for transposase accessible chromatin sequencing (scATAC-seq) is used to analyze chromatin accessibility and detect dynamic changes in chromatin structure at a single-cell level. Using scATAC-seq, Florian et al. revealed that chromatin structure is involved in the execution of different division modes of HSCs [37]. In details, they found that young HSCs mainly undergo asymmetric division, while old HSCs mainly undergo symmetric division. Furthermore, daughter cell potential correlates with the levels of the epigenetic mark H4K16ac and the amount of open chromatin allocated to daughter cells, suggesting that epigenetic mechanisms determine the functional outcome of HSCs division.

Single-cell high-throughput chromosome conformation capture (scHi-C) is the single-cell technology currently used to explore the three-dimensional (3D) structure of chromatin at a single-cell level. There has been a lack of complete studies on the 3D chromatin conformation of cell types at various stages of the hematopoietic lineage. The main limiting factor is that existing chromatin conformation capture techniques require large numbers of cells, and there are few HSCs in the body. Chen et al. introduced a low-input tag-based Hi-C method. They used this technique to generate 3D genomic data maps of 10 cell types during the differentiation of the mouse hematopoietic lineage. The development of scHi-C will help researchers understand various biological processes in the hematopoietic system [115].

PROTEOMIC ALTERATIONS CORRELATED WITH HEMATOPOIETIC AGING

Single-cell transcriptomics have transformed our understanding of hematopoietic system by facilitating gene expression measurements at high resolution, but protein-level expression measurements in single cells provide added value over measuring RNA-level expression alone, as the relationship between them is not a straightforward correlation for most genes. A few studies applied single-cell proteomics to the research in hematopoietic system. Palii et al. used single-cell proteomic with temporal barcoding to capture the temporal dynamics of protein expression of lineage-specific transcription factors (LS-TFs) in individual cells during human erythropoiesis using, and found that LS-TFs from alternate lineages are co-expressed in individual early progenitor cells and expression of them gradually changes to direct cell-fate decisions [116]. In a recent study, Zhang identified more than 80 different subpopulations of HSPCs by performing cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) on human bone marrow cells and constructed an immunophenotype-coupled transcriptomic atlas of human hematopoietic progenitors [40]. Zhao et al. adapted a high-throughput microfluidic-based technology to quantify 15 secreted proteins in HSPCs at single-cell level, and revealed that, upon Toll-like receptor stimulation, short-term HSCs and multipotent progenitor cells can produce a large number of diverse cytokines through NF-κB signaling [117]. Single-cell proteomics by mass cytometry time of flight (CyTOF) revealed impact of genetic alterations on frequency and lineage potential of HSC, providing mechanistic insights into altered hematopoiesis at single cell resolution [118]. However, few studies applied single-cell proteomics to the research in the aging dynamics of hematopoietic system, except that one group using single-cell proteomics by CyTOF measuring 20 surface proteins identified G6B as an indicator of mutant clone-derived HSPCs in myelofibrosis, which is often associated with aging [119].

METABOLOMICS ALTERATIONS CORRELATED WITH HEMATOPOIETIC AGING

The phenotypic manifestations in aging organisms are influenced by the repercussions of aging on physiological processes. A researcher performed proteomic analysis of human HSCs at different ages and other cells within the bone marrow niche. Their results show that the metabolic and anabolic activities of senescent hematopoietic progenitor cells (HPCs) are enhanced, and the lineage bias of HSPCs toward myeloid differentiation is related to glycolysis [35]. Studies have shown that aging phenotypes can be alleviated through caloric restriction [120]. Employing single-cell transcriptomic analysis on various tissues from calorically restricted aging rats uncovered the suppressive impact of caloric restriction on age-related inflammation [121].

While early studies tentatively proposed an association between dietary restriction (DR) and amelioration of HSC aging, these were primarily confined to phenotypic observations and lacked comprehensive exploration. Tao elucidated that intermediate-onset DR rectified the perturbed equilibrium in the HSC pool in aged mice, suppressing B lymphopoiesis under unperturbed conditions. Both short- and medium-term DR exhibited improvements in hematopoietic regeneration within aging HSCs, with long-term DR showcasing enhanced repopulation abilities of total bone marrow cells. Mechanistically, DR rejuvenated dysregulated mitochondrial pathways, increased quiescence and diminished DNA damage signaling [122]. Ma conducted extensive single-cell and single-nucleus transcriptome analyses across various tissues to investigate the impact of aging and caloric restriction in rats. The findings indicate that caloric restriction mitigates age-related alterations in cell type composition, gene expression and core transcriptional regulatory networks. This research offers a comprehensive single-cell transcriptional landscape across multiple tissues associated with mammalian aging and caloric restriction. It enhances our comprehension of caloric restriction’s resilience as an age-protective intervention and elucidates the mechanisms through which metabolic interventions reshape the aging process of the immune system. [121].

The gut microbiota stands as a pivotal regulator of host health and immunity. A recent study employing fecal microbiota transplantation (FMT) from young to aged mice unveiled noteworthy lymphoid differentiation in aged recipient mice, featuring increased and reduced myeloid differentiation. FMT from young mice exerted a rejuvenating effect on aged HSCs, enhancing both short- and long-term hematopoietic repopulation capacities. Metabolomic analysis disclosed the reshaping of intestinal microbiota composition and metabolites following FMT, with Treponema and tryptophan-related metabolites emerging as key contributors to hematopoiesis recovery and the rejuvenation of aged HSCs [123].

SINGLE-CELL AND SPATIAL MULTI-OMICS ANALYSIS OF HEMATOPOIETIC AGING

Single-cell techniques have advanced our understanding of hematopoietic aging, notably revealing many changes in aging HSCs (Fig. 2). Given the intricate landscape of cellular diversity within organisms, there arises a necessity for a comprehensive capture and analysis of multiple cellular processes to unravel this complexity. The advent of single-cell multi-omics technology addresses this need by facilitating the simultaneous integration of various single-omics methods, encompassing transcriptomics, genomics, epigenomics, epitranscriptomics, proteomics, metabolomics and other omics. This integrated approach elevates our comprehension of biological mechanisms and genotype–phenotype relationships, thereby instigating a transformative impact on molecular cell biology research [13, 124]. The evolution of single-cell multi-omics technologies empowers the direct characterization of phenotypes that had to be inferred through single-omics approaches. These advanced methodologies unravel the intricacies of gene regulation mechanisms, dynamics in protein expression, variations in epigenetic landscapes and models of cell perturbation. Consequently, they offer novel insights into comprehending cellular heterogeneity in both health and disease [15, 124–127], with a particular emphasis on studies within the hematopoietic system. At present, many achievements have been made in revealing the physiological or pathological processes of the hematopoietic system through Single-cell multi-omics methods. For example, deciphering cell status and human hematopoietic lineage [128, 129], explaining the occurrence and development of various blood system diseases [130–132]. CITE-seq addresses certain challenges inherent in single-cell analysis by integrating technologies for both transcriptomic and surface proteomic profiling. This innovative approach enables the concurrent detection of cell surface protein information and intracellular transcriptome information at the single-cell level. CITE-seq enhances the depth of understanding of individual cells, providing a more nuanced grasp of cellular heterogeneity and a more precise delineation of specific cell types, particularly within the intricate hematopoietic system [133–135]. However, it is essential to note that CITE-seq is not without limitations. It necessitates simultaneous antigen detection and transcriptome sequencing, imposing stringent requirements on cell viability for the latter process. Presently, CITE-seq predominantly targets membrane surface antigens to avoid potential impacts on the transcriptome when antibodies permeate the membrane. The extension of proteomic analysis tools using antibodies beyond cell surface proteins remains an area warranting further exploration and development.

The principal constraint of single-cell methodologies is rooted in their reliance on cell suspensions or isolated nuclei, necessitating subsequent disaggregation before analysis and consequently resulting in the forfeiture of crucial spatial information. The escalating demand for location-specific insights has spurred the development of spatiomics. As single-cell technology progresses, the capacity to investigate diverse molecular analytes within their native tissue context at subcellular resolution has given rise to the concept of ‘spatial multi-omics,’ acknowledged as a noteworthy technology by Nature magazine in 2022 [136]. Researches indicate that, beyond intrinsic factors, certain extrinsic elements can influence HSC aging, with the bone marrow microenvironment being a critical determinant of HSC function [137–145]. Moreover, non-hematopoietic cells play a crucial role in the HSC niche [146–148]. Spatiomics has proven valuable in diverse fields, yielding insights into development and brain structure [149–152], tumors [153–156] and neurodegenerative diseases [157, 158]. Hence, spatial multi-omics holds promise for elucidating the interplay between the aging and functional alterations of HSCs and the bone marrow milieu. Despite being in its nascent stages, spatial omics is advancing swiftly, and these nascent methodologies are poised to enhance our profound comprehension of biology [159].

FUTURE

Since Tang [11] pioneered the development of the world’s first single-cell sequencing technology in 2009, the field of single-cell technology has experienced rapid growth and widespread application over the past decade. Evolving from its initial capacity to analyze only a limited number of cells, the technology has progressed to simultaneously assess millions of cells. However, persistent challenges arise in ascertaining cellular origin and subsequent differentiation due to inherent technical constraints. To surmount these challenges, lineage tracing, at the forefront of technology, intervenes by imprinting cells with stable genetic markers, facilitating the tracking of cellular processes such as division, differentiation, migration and more. Wang have introduced a pioneering lineage tracing mouse model, DARLIN, consequently spearheading the development of Camallia-seq, the world’s premier single-cell multi-omics lineage tracing technology. This groundbreaking methodology allows for the concurrent detection of lineage markers, transcriptomic profiles, global DNA methylation and chromatin accessibility at the single-cell level. These innovations have proven pivotal in scrutinizing the fate selection of HSCs, their migration across diverse bone marrow microenvironments and the perpetuation of cellular molecular imprints. The significance of these advancements transcends individual studies, furnishing life sciences researchers with a potent tool to surmount challenges within their respective domains [160].

The landscape of technology is in constant flux to meet evolving demands, and the realm of single-cell technology is no exception, experiencing continuous innovation. Specifically, second-generation sequencing-based scWGS has the capability to interrogate genetic variations at the whole-genome level at the single-cell resolution. However, the relatively short sequencing read length associated with scWGS introduces certain limitations. In contrast, third-generation sequencing (TGS) technology presents notable advantages in examining the transcriptome of individual cells. The strengths of TGS technology lie in its long-read lengths, elimination of the need for amplification, facilitation of direct sequencing, and minimal input requirements. TGS technology has emerged as an indispensable tool for single-cell analysis, providing more exhaustive, accurate, and profound single-cell transcriptome data.

Despite the widespread adoption of single-cell omics in both basic research and clinical trials, several notable challenges persist. The preparation of single-cell samples typically entails compromises in terms of cell integrity and activity. The enhancement of preparation methods holds the potential to improve both cell throughput and accuracy. Determining the optimal sequencing depth is another intricate challenge, as heightened depth escalates sequencing costs. Achieving a judicious equilibrium between throughput, cost considerations, and the intricacies of data processing is imperative. This entails a comprehensive evaluation of research objectives, available resources and pertinent technical attributes.

To comprehensively elucidate the molecular hierarchy from the genome to phenotype within a single cell, the integration of multi-omics approaches with both single-cell and spatial resolution becomes imperative. These techniques enable a clear exploration of the intermolecular dynamics between gene regulation and gene expression within the same cell type, particularly in investigations pertaining to development, aging and disease. Furthermore, these advanced technologies empower the investigation of how acquired genetic variations in single-cell genomes influence their functional and phenotypic characteristics, as well as the functions of surrounding tissues. This not only provides avenues for a broad range of applications in cell biology but also offers unprecedented insights into the intricacies of biological processes. The ongoing progress in single-cell technology and the convergence of multi-omics with spatial resolution are poised to drive further innovation, promising even more comprehensive insights into our understanding of biology in the future.

ACKNOWLEDGEMENTS

This work was supported by the National Key R&D Program of China (2021YFA1102800) and by National Natural Science Foundation of China (92249304, 82301746). This work was supported by the Science Foundation for Distinguished Young Scholars of Guangdong Province (2019B151502008) to Hu Wang. This work was also supported by the Hangzhou Youth Innovation Team Project (TD2023020).

AUTHOR CONTRIBUTIONS

Xinrong Jin (Investigation [equal], Software [equal], Visualization [equal], Writing— original draft [lead), Ruohan Zhang (Investigation [equal], Writing—original draft [supporting]), Yunqi Fu (Investigation [equal], Writing—original draft [supporting]), Qiunan Zhu (Investigation [equal], Writing—original draft-Supporting), Liquan Hong (Conceptualization [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing—review & editing [equal]), Aiwei Wu (Software [equal], Supervision [equal], Visualization [equal], Writing—review & editing [equal]), Hu Wang (Conceptualization [equal], Funding acquisition [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing—review & editing [equal]).

Author Biographies

Xinrong Jin is currently pursuing his Ph.D. in Biology from Hangzhou Normal University under the supervision of Hu Wang. His research interests focus on the mechanism of stem cell aging.

Ruohan Zhang is an undergraduate student at the Basic Medical Sciences at Hangzhou Normal University. Her research interests focus on mechanism of stem cell aging.

Yunqi Fu is an undergraduate student at the Basic Medical Sciences at Hangzhou Normal University. His research interests focus on machine learning for data processing.

Qiunan Zhu is an undergraduate student at the Basic Medical Sciences at Hangzhou Normal University. His research interests focus on the mechanism of mouse adipocytes aging.

Liquan Hong, is the director of the Third People’s Hospital of Deqing, China. His research interests focus on the mechanism of stem cell aging.

Aiwei Wu earned her Ph.D. from Shanghai Jiao Tong University. Her research interests lie in understanding the mechanisms of hematopoietic aging utilizing cutting-edge single-cell multiomics and bioinformatics approaches.

Hu Wang earned his Ph.D. from Zhejiang University. He is currently dedicated to investigating the molecular mechanisms of adult stem cell aging, as well as advancing the development and application of anti-aging drugs.

References

Author notes