Aging mechanisms and rejuvenation strategies for hematopoietic stem cells

Abstract

Hematopoietic stem cells (HSCs) transition through different functional states throughout life from emergence and expansion in the fetus, homeostasis maintenance in adulthood, and progressive functional decline with age. Aged HSCs are characterized by increased phenotypic number, decreased self-renewal and long-term reconstitution capacity, myeloid-biased differentiation, and clonal hematopoiesis. In this review, we summarize the life cycle of HSCs, integrate recent advances in understanding the cell-intrinsic and extrinsic mechanisms that drive HSC aging, and highlight innovative rejuvenation strategies that could be harnessed to delay HSC and systemic aging.

Introduction

HSCs drive lifelong blood production by balancing self-renewal and multilineage differentiation [1], which undergo progressive functional decline during aging. Key hallmarks of HSC aging include expansion of the HSC pool, decreased self-renewal and long-term reconstitution, myeloid-biased differentiation, as well as CH and increased susceptibility to oncogenic transformation [2, 3]. These age-related alterations not only disrupt hematopoietic homeostasis, but also lead to geriatric anemia, myelodysplastic syndromes (MDS), leukemia or immunodeficiency disorders [4]. Notably, HSC aging is a key driving force of systemic aging that is characterized by multi-organ inflammation and dysfunction. Therefore, dissecting the mechanisms by which HSCs age and developing novel rejuvenation strategies accordingly are of great significance to both hematopoietic and systemic aging.

HSC aging is regulated by both cell-intrinsic and extrinsic factors in the BM hematopoietic microenvironment (HSC niche) [5]. Major intrinsic alterations associated with HSC aging include DNA damage [6, 7], telomere shortening [8], epigenetic drift [9], impaired proteostasis and autophagy [10, 11], as well as mitochondrial dysfunction [12]. As HSCs age, the BM microenvironment also undergoes dramatic changes such as pro-inflammatory factor accumulation [13], increased adipogenesis and decreased osteogenesis [14], diminished arterioles and sympathetic innervation [15], sinusoid dilation and leakiness [16], reduced blood flow [17], BM fibrosis [18], as well as increased matrix stiffness [19]. However, which of these alterations could be effectively targeted to delay or reverse HSC aging is still poorly understood.

Given that HSC aging is pivotal to hematopoietic and systemic aging, developing novel HSC rejuvenation strategies should not only treat blood disorders, but also contribute to systemic anti-aging. Current HSC rejuvenation strategies include physical exercise [20, 21], dietary/calorie restriction [22], heterochronic parabiosis [23], genetic manipulation and cell reprogramming [24–28], pharmacological interventions [12, 29–34], clearing of aged HSCs [35, 36], and ex vivo reconstitution of young BM niche [37]. In this review, we will summarize the aging hallmarks of HSCs, their aging mechanisms and rejuvenation strategies, and propose potential future directions in this exciting research area.

Life cycle of hematopoietic stem cells

HSC emergence during embryonic development

Formation of the vertebrate hematopoietic system proceeds through a tightly coordinated spatiotemporal process, evolving from primitive to definitive hematopoiesis in three major waves [38]. In the first wave, primitive hematopoiesis initiates in the mouse yolk sac at around mouse embryonic day 7 (E7) [39]. In the second wave (pro-definitive hematopoiesis), erythro-myeloid progenitors (EMPs) and lymphoid-myeloid progenitors arise in the mouse yolk sac at around E8.25-E9, which support transient blood supply and generate tissue-resident macrophages and innate lymphoid cells in adult mice [40]. In the third wave, definitive hematopoiesis, characterized by HSC emergence and multi-lineage differentiation, occurs around E10.5-E11.5 in the aorta-gonad-mesonephros (AGM) region [41, 42]. Following their entry into the circulation around E11.5 and subsequent colonization of the fetal liver, HSCs undergo numerical expansion, peaking at approximately E14.5 [43]. While primitive and pro-definitive hematopoiesis support embryonic development, definitive hematopoiesis is indispensable for lifelong blood production and immune function (Fig. 1).

Fig. 1.

The life cycle of HSCs. The life cycle of HSCs can be divided into four key phases. HSC emergence at the embryonic stage: Primitive hematopoiesis in the yolk sac generates primitive erythrocytes and macrophages, followed by a pro-definitive stage marked by the emergence of EMPs. Definitive hematopoiesis is initiated by the emergence of self-renewing, multipotent HSCs in the AGM region. These HSCs subsequently migrate to and expand in the fetal liver, supporting fetal blood production; HSC transition at the perinatal stage: Around birth, HSCs migrate from the fetal liver to the BM where they undergo a critical transition from active cycling to a quiescent state; HSC maintenance at the adult stage: In adulthood, HSCs achieve a homeostatic balance. They predominantly maintain a quiescent state while staying poised to meet physiological demands, thereby ensuring lifelong hematopoiesis. HSC aging at the final stage: With age, the hematopoietic system exhibits signs of functional exhaustion, characterized by increased HSC number, decreased self-renewal capacity, a bias toward myeloid differentiation and lost of HSC polarity. Figure created by BioRender

Perinatal transition and adult maintenance

The perinatal period constitutes a pivotal transition for HSCs, which migrate from the fetal liver to the BM and initiate a fundamental shift in their functional state [44]. HSCs transition from an active cycling state to a predominantly quiescent state within 4 weeks after birth [45], which is coordinately driven by intrinsic and extrinsic factors.

Upon reaching adulthood, HSCs are balanced between long-term stem cell pool maintenance and efficient responsiveness to the body's dynamic blood demand. This is accomplished through a multi-layered, highly coordinated regulatory network. Under homeostatic conditions, HSCs undergo infrequent divisions to achieve self-renewal, thereby sustaining stable numbers and functionality over time [46]. Their central physiological role is the continuous production of a balanced and diverse repertoire of mature blood cell lineages. Although the majority of HSCs are maintained in quiescence (G0 phase) during homeostasis [46], they can be effectively activated upon exposure to stresses such as acute infection, hemorrhage, chemotherapy, or inflammatory signals (e.g., type I interferons), thereby prompting cell cycle entry, rapid proliferation, and differentiation to replenish blood cells [47]. Once the stress resolves, negative feedback mechanisms enable HSCs to return to quiescence [46] (Fig. 1).

Functional decline with age

During aging, the HSC regulatory network undergoes stepwise failure [48]. A notable paradox is the coexistence of declining self-renewal capacity with an expansion in phenotypically defined HSC numbers [49]. Serial transplantation assays reveal that aged HSCs produce progeny clones of reduced size that are prone to exhaustion, reflecting impaired proliferative sustainability [50]. Furthermore, aged HSCs exhibit marked lineage bias, characterized by enhanced myeloid output and a sharp decline in lymphopoietic capacity [51]. Two models have been proposed to explain this phenomenon: one posits a homogeneous functional shift across all HSCs, leading to systemic bias toward myeloid output [51]; the other suggests an alteration in HSC subtypes, with a reduction in lymphoid-biased (Ly-bi) HSCs and accumulation of myeloid-biased (My-bi) HSCs [52].

Beyond lineage bias, the balance between symmetric and asymmetric division is also disrupted during HSC aging. Young HSCs maintain a tightly regulated equilibrium among symmetric self-renewal, asymmetric division, and symmetric commitment [53, 54]. With aging, HSCs exhibit a marked shift toward symmetric commitment divisions at the expense of asymmetric divisions, leading to accelerated depletion of the stem cell pool [53, 55]. This shift is closely associated with loss of cellular polarity, as evidenced by redistribution of polarity markers such as Cdc42 and tubulin. Aged HSCs also display reduced homing efficiency due to decreased expression of homing receptors such as CXCR4 [56, 57]. Deterioration of the BM microenvironment also disrupts HSC-niche communication, further accelerating hematopoietic aging [58] (Fig. 1).

It is intriguing to note that HSC aging proceeds in an inverted sequence of its developmental trajectory, characterized by diminishing lymphopoietic potential (i.e. compromised immune system), as well as increased myelopoiesis and thrombopoiesis that are reminiscent of primitive hematopoiesis. Furthermore, CH is also reminiscent of the onset of definitive hematopoiesis where a few HSC clones dominate the blood system. In summary, HSCs undergo dramatic changes throughout their lifecycle, which are driven by a series of highly conserved cellular and molecular events. In the following section, we will focus on the key mechanisms by which aging impairs HSC function.

Cellular and molecular mechanisms underlying HSC aging

HSC aging results from a complex interplay between cell-intrinsic alterations and extrinsic influences from the aging BM. Key intrinsic alterations include DNA damage, telomere shortening, epigenetic drift, dysregulated proteostasis and autophagy, and mitochondrial dysfunction. Deterioration of the BM niche also accelerates HSC aging [59–62]. This section summarizes recent advances in both cell-intrinsic and cell-extrinsic factors that drive HSC aging (Fig. 2).

Fig. 2.

Cell-intrinsic and extrinsic alterations during HSC aging. The functional decline of HSCs with age arises from both cell-autonomous defects and a deteriorating BM microenvironment. Intrinsically, aged HSCs accumulate cellular damage that compromises their function, which include DNA damage and telomere attrition that cause genomic instability, and epigenetic drift that alters transcriptional programs. Furthermore, disturbances in proteostasis, altered autophagy and mitochondrial dysfunction contribute to functional decline of HSCs by impairing protein quality control and organelle turnover. Extrinsically, the aged BM microenvironment exhibits a marked shift in cellular composition and ECM properties, characterized by depletion of key supportive stromal cells, accumulation of pro-inflammatory factors, adipocyte accumulation, and BM stiffening. Vascular degeneration such as arteriole shortening, sympathetic denervation, sinusoid dilation and leakage also occur with age. Together, these multifaceted intrinsic and extrinsic alterations drive the characteristic aging phenotypes of HSCs. Figure created by BioRender

DNA damage and telomere shortening

DNA damage accumulation is the pivotal driving force of HSC aging [63]. As a hallmark of cellular senescence, DNA damage arises from diverse physical, chemical and biological stressors, which drive genomic instability and induce apoptosis or senescence [64]. Beerman et al. reported that age-associated DNA damage preferentially accumulates in HSCs [65]. DNA damage-induced nuclear translocation of MCPH1 activates necroptosis pathways and promotes HSC aging [66]. Gadd45a is a key regulatory protein in response to DNA damage, whose expression gradually decreases with age [67, 68]. In young HSCs, Gadd45a deficiency leads to delayed DNA damage repair and accumulation of intracellular damage. Suppression of apoptotic pathways enables damaged HSCs to survive in the short term, which exhibit enhanced regenerative capacity and stress resistance. However, continuous accumulation of DNA damage significantly increases the risk of malignant transformation in HSCs, ultimately predisposing them to the development of B-cell leukemia [69]. The circadian protein Per2 is also activated during DNA damage and aging, particularly in lymphoid-biased HSCs [70]. Per2 amplifies DNA damage signaling and activates p53-dependent apoptotic pathway, leading to selective depletion of lymphoid-biased HSCs, aging-related lymphopenia and immune dysfunction [70]. Deletion of Per2 alleviates apoptosis and improves immune function during aging, without increasing DNA damage or cancer risk [70].

Telomere attrition serves as a fundamental molecular mechanism underlying cellular senescence. Once telomeres are shortened to a critical threshold, cells enter a senescent state characterized by irreversible cell cycle arrest, leading to diminished tissue regeneration [71]. HSCs in elderly individuals exhibit significantly shorter telomere lengths [72]. In HSCs, telomere attrition markedly impairs proliferation and differentiation, thereby compromising their ability to maintain hematopoietic homeostasis [73, 74]. Taken together, age-related functional decline of HSCs likely arises from both cumulative DNA damage and telomere shortening.

Epigenetic changes during HSC aging

DNA methylation

Mammalian genomes encode three DNA methyltransferases (e.g., DNMT1 for methylation maintenance, DNMT3A and DNMT3B for de novo methylation) that govern HSC homeostasis [6, 75–77]. Genetic deletion of Dnmt1 induces myeloid-erythroid bias coupled with lymphoid depletion in mice [7, 78]. Beerman et al. showed that age-associated DNA methylation remodeling in HSCs preferentially targets loci governing self-renewal capacity and lineage commitment [79]. Mutations of the DNA demethylase TET2 also disrupt lineage-specific transcription factor (TF) motif methylation, and induce myelomonocytic skewing [80]. Functional restoration of TET2 in mutant leukemia cells can block aberrant self-renewal and delay disease progression by rectifying the epigenetic landscape [81, 82]. Notably, DNMT3A and TET2 exhibit antagonistic roles in DNA methylation and lineage commitment yet converge to restrict HSC self-renewal [83]. This functional dichotomy underscores the necessity for a fine-tuned equilibrium between DNA methylation and demethylation in maintaining hematopoietic homeostasis.

DNMT3A and TET2 are frequently mutated in age-related CH [84, 85]. However, the relationship between CH and HSC aging is complex. Whereas both mutant and wild-type HSCs from CH individuals show enhanced inflammatory signatures versus non-CH individuals, mutant HSCs exhibit an "aging-resistant" transcriptional signature with attenuated inflammatory responses compared to wild-type HSCs within the same microenvironment [86]. These findings propose a refined clonal selection model in CH: mutant clones are favored not because they drive aging, but because they resist the inflammatory milieu that suppresses wild-type HSCs. Thus, the "aging-resistant" phenotype paradoxically fuels clonal expansion, which, at the organismal level, is associated with increased risks of blood cancer [87] and cardiovascular disease [88].

Histone modifications

Histone modifications represent fundamental epigenetic mechanisms governing gene expression [89]. These chemical modifications, including acetylation, methylation, phosphorylation, sumoylation, and ubiquitination, can modulate transcriptional activity through chromatin architectural reorganization and DNA accessibility control [90, 91]. Emerging evidence underscores the critical roles of histone modifications in governing stem cell self-renewal and lineage commitment [92–94]. Sun et al. showed age-associated elevation of H3K4me3 levels that strongly correlates with transcriptional dysregulation by analyzing young versus aged HSCs [95]. Deletion of the deacetylase Sirt1 from young HSCs induces myeloid-biased differentiation accompanied by decreased lymphoid compartment, anemia, and transcriptional reprogramming [96]. Aged LT-HSCs exhibit diminished H4K16ac levels compared to young counterparts, together with loss of polarity during hematopoietic aging [29]. Notably, age-related chromatin alterations seem to be reversible, since decreasing Cdc42 activity can restore a youthful epigenetic landscape in aged HSCs [97].

Noncoding RNAs

Non-coding RNAs (ncRNAs) are key regulators of HSC biology, controlling processes like self-renewal, lineage differentiation, and homeostasis [98, 99]. miRNAs are essential for HSC homeostasis, and their disruption can trigger blood disorders [100]. During aging, elevated miR-125b directly targets and represses the histone methyltransferase SUV39H1, leading to impaired B cell differentiation [101]. Conversely, inhibiting miR-125 or restoring SUV39H1 in aged HSCs rescues their B cell differentiation potential [101]. Similarly, the miR-212/132 cluster (Mirc19) is also upregulated with age, which promotes HSC exhaustion by cell cycle acceleration via targeting the pro-longevity factor FOXO3 [102]. Therefore, miR-125b and miR-212/132 emerge as critical, pathway-specific regulators of epigenetic and cell cycle control, whose dysregulation drives distinct aspects of HSC functional decline during aging.

Proteostasis and autophagy

Maintaining proteostasis in HSCs is crucial for their self-renewal and maintenance [103]. With aging, the proteostasis network in HSCs becomes progressively dysregulated. Aged HSCs not only produce more misfolded proteins but also exhibit a concomitant decline in proteasome-mediated degradation, leading to accumulation of protein aggregates that cause premature exhaustion of the HSC pool and functional decline [104]. Autophagy, a lysosomal degradation pathway responsible for clearing damaged cellular components, is also impaired in HSCs during aging [11]. Declined autophagic activity contributes to a substantial increase in reactive oxygen species (ROS) production, exacerbating oxidative stress and disrupting the quiescent state of HSCs. This leads to loss of self-renewal capacity and emergence of aging-associated phenotypes, such as myeloid-biased differentiation [105]. Consistent with this, conditional deletion of Atg7 severely impairs autophagy in HSCs, which causes mitochondrial dysfunction, accumulation of ROS, and triggers abnormal cell proliferation and DNA damage [106]. Consequently, the number of functional HSCs is significantly reduced, while the generation of both lymphoid and myeloid progenitors is severely suppressed [106]. Dellorusso et al. demonstrated that inflammatory signals suppress glycolysis in old HSCs through Socs3-mediated inhibition of Akt/FoxO pathway, leading to reduced glucose uptake and metabolic stress [107]. In response to this glycolytic impairment, HSCs engage autophagy as an adaptive cytoprotective mechanism to preserve functional quiescence. Remarkably, transient autophagy induction via a short-term fasting/refeeding paradigm normalizes glycolytic flux and significantly enhances the regenerative capacity of aged HSCs in transplantation assays [107].

Mitochondrial dysfunction

Mitochondrial injury is a hallmark of aged HSCs. For instance, clusterin is upregulated in aged HSCs and promotes mitochondrial hyperfusion through its interaction with Mfn2. Clusterin deficiency enhances mitophagy and reverses the myeloid differentiation bias [108]. Mitochondrial dysfunction is also reflected in the heterogeneity of mitochondrial membrane potential (MMP), which reduces in a subset of aged HSCs, while high-MMP cells retain youthful transcriptional features [12]. The correlation between MMP and functional fitness suggests that targeting MMP could be a feasible rejuvenation strategy. Mechanistically, this MMP heterogeneity and structural injury stem from impaired mitochondrial quality control, as mitophagy is compromised in aged HSCs, leading to accumulation of dysfunctional organelles [109]. Mitochondrial stress further activates the NLRP3 inflammasome in aged HSCs via SIRT2 downregulation, thereby establishing a vicious cycle of damage and inflammation [110].

Mitochondrial injury inevitably leads to leakage of mitochondrial contents into the cytosol, a phenomenon increasingly recognized as a driver of cellular aging but remains largely unexplored in HSCs. Currently, the most definitive evidence of mitochondrial content leakage in HSCs comes from ROS [111]. ROS play a dual role in HSC biology. At low levels, they are required as signaling molecules to maintain quiescence and self-renewal, whereas excessive accumulation constitutes a hallmark of HSC aging and a key driver of functional decline [112]. Young quiescent HSCs predominantly utilize anaerobic glycolysis with minimal mitochondrial OXPHOS, a metabolic adaptation that limits ROS production and preserves stemness [113, 114]. In contrast, aged HSCs exhibit a metabolic shift toward enhanced OXPHOS, resulting in elevated ROS production [25, 115]. This pro-oxidant state is further exacerbated by the decline of antioxidant defenses, notably selenoproteins like GPX4, which fail to neutralize ROS and lipid peroxides in aged HSCs [116]. The cytosolic ROS subsequently cause DNA damage, lipid peroxidation, and protein oxidation, ultimately inducing premature aging phenotypes in HSCs, such as diminished self-renewal capacity, myeloid differentiation bias, and impaired reconstitution ability after transplantation. Concurrently, aberrant opening of the mitochondrial permeability transition pore (mPTP) is a key mechanism driving ROS leakage [117]. Studies have found that Nynrin deficiency leads to excessive mPTP opening, which not only causes mitochondrial swelling but also promotes the release of ROS into the cytosol [118].

In summary, mitochondrial dysfunction in aged HSCs manifests as structural injury, content leakage and oxidative stress. These interconnected defects form a self-reinforcing cycle that drives HSC aging.

Alterations in the extrinsic HSC niche

Age-related remodeling of the BM niche significantly impairs its hematopoietic supporting capacity. Key alterations include decreased bone volume, increased marrow adiposity, sympathetic denervation, elevated vascular permeability, and increased matrix stiffness, which synergistically compromise HSC functionality [15, 59, 119] (Fig. 2).

Bone marrow stromal cells

LepR⁺ BM stromal cells (BMSCs) and endothelial cells (ECs) constitute the most important HSC-supporting cells within the perivascular niche [120–123]. Mo et al. found that in young mice, LepR⁺ BMSCs express high levels of HSC niche factors (e.g., Kitl, Cxcl12) and predominantly differentiate into osteolineage cells under steady state. In contrast, aged mice show a shift of LepR⁺ BMSCs toward adipogenic lineage [124], which is accompanied by decreased osteoblast numbers and downregulation of osteogenic and HSC niche factors [33, 125]. Young et al. showed that HSC aging is linked to declining IGF1 levels in the BM [125]. However, Igf1 deletion from BMSCs using Lepr-Cre did not induce premature HSC aging [126]. Osteolectin⁺ (Oln⁺) cells represent a distinct periarteriolar subpopulation of LepR⁺ BMSCs that function as osteolineage-biased progenitors [15]. Importantly, these cells also create a specialized niche via Piezo1-mediated mechanosensing and SCF expression to support common lymphoid progenitors (CLPs) [127]. During aging, this periarteriolar niche undergoes significant deterioration, characterized by reduced Oln⁺ cell frequency and subsequently diminished lymphopoiesis [127]. Importantly, BM stiffening was also found to be a novel hallmark of hematopoietic aging, which down-regulates HSC niche factors in BMSCs and drives HSC aging [19].

Osteoblasts

Osteoblasts are also critical cellular components of the HSC niche [128]. Aging leads to a significant reduction in osteoblast numbers in the BM, primarily due to impaired osteogenic differentiation by BMSCs and increased apoptosis [129, 130]. During aging, key osteogenic transcription factors such as Runx2 and Dlx5 are significantly downregulated, causing reduced bone formation. In contrast, age-related RANKL/OPG imbalance promotes bone resorption by increasing osteoclast activity [130]. A subset of osteoblasts enters cellular senescence, acquiring a senescence-associated secretory phenotype (SASP) [131]. Mitochondrial dysfunction and elevated oxidative stress further compromise their function [132]. These changes not only contribute to osteoporosis, but also impair HSC support by disrupting the BM microenvironment [128, 133]. Targeting aged osteoblasts by senolytic clearance, restoring osteo-adipogenic balance, or improving mitochondrial function may offer novel strategies to rejuvenate the aged marrow niche and delay hematopoietic and immune aging [131, 132].

Adipocytes

Adult BM also harbors adipocytes, the frequency of which is inversely correlated with the hematopoietic activity under steady state [62, 134–136]. Aging leads to a significant increase in BM adipocytes, a phenomenon known as "bone marrow yellowing"—one of the most prominent features of marrow aging. The number of adipocytes in the proximal tibia is markedly higher in 26-month-old aged mice than in 8-month-old adult mice [137]. More importantly, aging profoundly reshapes their functional characteristics. The expression of the adipogenic transcription factor PPAR-γ2 increases, while osteogenic regulators Runx2 and Dlx5 decline [137]. Epigenetically, miR-188 is upregulated to promote adipogenesis, and to suppress osteogenesis by targeting HDAC9 and RICTOR [129]. This functional remodeling of aged BM adipocytes not only alters the local microenvironment but also influences systemic health through multiple mechanisms. Their pro-inflammatory secretome skews HSC differentiation toward the myeloid lineage at the expense of lymphopoiesis, thereby impairing adaptive immunity [130]. Under stress conditions such as chemotherapy or transplantation, aged BM adipocytes exhibit a reduced capacity to support HSCs, compromising hematopoietic recovery [128, 130]. Targeting adipocyte-derived factors or modulating β-adrenergic signaling may help establish a more youthful marrow niche and improve hematopoietic and immune function in the elderly [130].

BM vasculature

With aging, the BM vasculature undergoes profound functional and structural changes [138]. These include arteriole shortening and sympathetic denervation [15], sinusoid dilation and leakage [59], reduced blood flow [17, 139], and elevated ROS production [37]. The age-related vasculature remodeling coincides with a functional decline of perivascular BMSCs and osteoprogenitors [59, 138]. When young HSCs are co-cultured with aged ECs or transplanted into recipients conditioned by aged ECs, their long-term repopulating capacity is significantly compromised, accompanied by a pronounced myeloid bias in their lineage output. In contrast, young ECs not only enhance the regenerative potential of aged HSCs but also accelerate hematopoietic recovery following myelosuppressive injury [37].

Megakaryocytes

Megakaryocytes closely associate with HSCs to promote their maintenance and regeneration by secreting TGFβ1 and CXCL4 [140, 141]. Interestingly, the frequency and number of BM megakaryocyte progenitors are significantly higher in aged mice [142]. Aging activates direct differentiation from HSCs to megakaryocytes, leading to increased thrombosis [143]. The expression of PF4 (CXCL4) in aged megakaryocytes is decreased, and supplementation with PF4 can partially reverse HSC aging [144]. Single-cell RNA-sequencing (scRNA-seq) reveals abnormal expression of pathways related to mitochondria and inflammation in aged megakaryocytes [145]. These findings suggest that paracrine factors secreted by megakaryocytes may serve as potential targets for rejuvenating HSCs and preventing aging-related diseases.

Sympathetic nerves

The nervous system also play a crucial role in regulating bone and BM function [146]. While parasympathetic nerves primarily innervate bone tissue without penetrating the marrow, sympathetic and sensory nerves innervate both compartments [146, 147]. The sympathetic nervous system (SNS) has been identified as a key regulator of HSC aging [15]. Loss of sympathetic innervation or disruption of adrenergic receptor beta-3 (ADRβ3) signaling is sufficient to induce HSC aging [15]. This aging phenotype is driven by remodeling of the BM microenvironment, characterized by shortening of arterial vessels and downregulation of critical HSC maintenance factors (e.g., CXCL12 and SCF) in BMSCs [15].

Macrophage and regulatory T cells

Chronic inflammation within the BM microenvironment is a key driver of HSC aging. Macrophages [148] and regulatory T cells (Tregs) [149], which are niche components that maintain HSC homeostasis in young BM, undergo a functional shift in the context of aging. Aged macrophages exhibit an activated phenotype with increased inflammatory signaling, elevated IL-1β levels, and enhanced caspase-1 activity [150]. Importantly, the phagocytic function of aged macrophages is severely impaired, rendering them unable to effectively clear aged neutrophils, leading to accumulation of these cells in the BM. The uncleared senescent cells release damage-associated molecular patterns, which further amplify inflammatory signals, forming a vicious cycle [150]. Unlike macrophages that directly secrete pro-inflammatory cytokines, aged Tregs contribute to inflammation primarily through functional impairment. The age-related expansion of a dysfunctional KLRG1⁺ Treg subset, coupled with their inherent dependence on mitochondrial oxidative phosphorylation, renders them vulnerable to mitochondrial damage and unable to effectively suppress inflammatory responses, thereby perpetuating immune dysregulation in the aging BM [151].

Biomechanical regulation

While most HSC niche studies are focused on biochemical signals, a growing body of evidence highlights the critical role of biomechanical forces in regulating HSC fate throughout life [152, 153]. These mechanical cues, including shear stress from blood flow, extracellular matrix (ECM) stiffness and cyclic strain, are transduced into biochemical signals through mechanosensitive pathways, a process known as mechanotransduction [154, 155]. The importance of biomechanics is perhaps most evident during embryonic HSC specification. Studies in zebrafish and mouse models have demonstrated that hemodynamic forces generated by blood flow are essential for endothelial-to-hematopoietic transition (EHT) in the AGM region, where the first definitive HSCs emerge [156, 157]. Mechanical forces trigger nuclear and mitochondrial adaptations, regulating cell differentiation [158]. These developmental insights suggest that biomechanical cues could influence HSC behavior throughout life. Importantly, HSC niche within the BM microenvironment (diaphyseal region of limb bones) significantly stiffens with age, as measured by atomic force microscopy (AFM) [19]. BM stiffening activates nuclear translocation of mechanotransducers YAP and TAZ in BMSCs, which suppresses SCF/CXCL12 secretion and promotes HSC aging [19]. However, the cellular and molecular mechanisms underlying BM stiffening remain to be explored, which could be attributed to BM fibrosis or vascular sclerosis induced by chronic inflammation.

HSC rejuvenation strategies

Based on the cell-intrinsic and extrinsic mechanisms that drive HSC aging, both systemic and targeted approaches have been developed to promote HSC rejuvenation, which we summarize as below (Fig. 3 and Table 1).

Fig. 3.

Current strategies for HSC rejuvenation. Multiple approaches have been applied to promote HSC rejuvenation, targeting either intrinsic HSC alterations or extrinsic BM microenvironment. Figure created by BioRender

Table 1.

Summary of HSC rejuvenation strategies

Strategies	Intervention	Phenotype	Refs
Exercise	Exercise	No change of HSC number or CD150+ cell frequency	[160]
		Expanded lymphoid progenitors and LepR+Oln+ periarteriolar stromal cells	[127]
Dietary restriction and metabolic intervention	30% DR	Preserved HSC regeneration but impaired lymphopoiesis with DR	[22]
	Prolonged fasting	HSC rejuvenation via prolonged fasting	[161]
	NR	Restored metabolism and rejuvenated HSCs in aged mice with chronic NR treatment	[162]
	mTOR inhibition	Restored self-renewal, reconstitution and immunity with rapamycin	[163]
	Microbiota	HSC rejuvenation via young microbiota transplantation	[164]
	Uridine	Enhanced HSC self-renewal and suppressed inflammation via uridine supplementation	[165]
Genetic manipulation and cellular reprogramming	Sirt3	Enhanced regenerative capacity of aged HSCs with Sirt3 overexpression	[24]
	Sirt7	Reversed aging hallmarks with Sirt7 overexpression	[25]
	Satb1	Enhanced lymphopoiesis with Satb1 overexpression	[27]
Small-molecule drugs or protein factor therapy	Casin	Restored youthful HSC polarity	[29]
		Lifespan extension and anti-inflammation	[30]
		Partial functional restoration of human HSCs by ex vivo Casin treatment	[31]
	Mito-Q	Enhanced mitochondrial function	[12]
	OPN	Partial polarity reversal by OPN treatment in vitro	[32]
	IGF1	Reduced aging hallmarks and restored mitochondrial function with exogenous IGF1 treatment	[33]
Senolytic drugs and targeted removal of aged HSCs	ABT263	Senolytic clearance of aged HSCs to enhance HSC repopulation with ABT263	[34]
	Myeloid-biased HSC depletion	CD150low HSCs can extend the lifespan of old mice	[35]
	Anti-CD150 antibody	Increased lymphopoiesis with anti-CD150 antibody treatment	[36]
Heterochronic parabiosis	Parabiosis	Transcriptional but not epigenetic or functional rejuvenation of HSCs	[160, 171, 172]
Reconstitution of young BM niche	3D co-culture of BMSCs and HSCs in GelMA	Increased HSC polarity, MMP, lymphopoiesis, and multilineage reconstitution capacity	[19]

Exercise

Exercise enhances systemic health through improved circulation and metabolism [159]. However, Ho et al. found that exercise has little effect on rejuvenating HSCs [160]. In contrast, Shen et al. demonstrated that physical exercise in 18-month-old mice significantly increases Lepr⁺Oln⁺ perivascular niche cells and expands lymphoid progenitor populations [127]. It is plausible that exercise preferentially accelerates lymphopoiesis via niche remodeling rather than directly rejuvenates aged HSCs.

Dietary restriction and metabolic intervention

Dietary restriction (DR) modulates HSC function through multiple nutrient-sensing pathways, including IGF-1, mTOR, AMPK, and NAD⁺/Sirtuin signaling. Tang et al. demonstrated that chronic 30% DR preserves the regenerative capacity of HSCs in aged mice, albeit at the cost of impaired lymphoid differentiation due to suppressed progenitor proliferation [22]. Cheng et al. found that prolonged fasting rejuvenates HSCs through coordinated IGF-1 reduction and PKA inhibition [161]. In contrast, Ho et al. reported that lifelong calorie restriction (CR) fails to improve the function of aged HSCs [160].

Pharmacological approaches targeting metabolic pathways also show rejuvenation effects on HSCs. Nicotinamide riboside (NR) enhances mitochondrial function and restores the metabolic competence and regenerative capacity of aged HSCs [162]. Similarly, mTOR inhibition with rapamycin reverses age-related functional decline of HSCs, improving self-renewal, reconstitution potential, and antiviral immunity in aged mice [163]. Interestingly, transplantation of young microbiota into aged mice could also rejuvenate HSCs in a systemic metabolite-dependent manner [164]. Uridine, a metabolite enriched in young hematopoietic cells, can also rejuvenate aged HSC by enhancing self-renewal and suppressing inflammation through FoxO singling pathway [165]. Together, these interventions demonstrate that metabolic reprogramming can partially delay hematopoietic aging, though with varying efficacy across experimental paradigms.

Genetic manipulation and cellular reprogramming

Sirt3 is a mammalian deacetylase that exhibits age-dependent expression decline in HSCs. Brown et al. demonstrated that Sirt3 knockout in aged mice impairs HSC self-renewal capacity, while its overexpression enhances regenerative potential [24]. Similarly, Mohrin et al. revealed that Sirt7 deletion induces premature activation, lymphoid differentiation bias, and functional exhaustion of HSCs. In contrast, Sirt7 overexpression reverses these aging hallmarks, and restores balanced lineage output and reconstitution capacity in geriatric murine models [25]. Satb1 is a chromatin remodeling regulator critical for lymphoid progenitor specification [166]. Satoh et al. demonstrated that Satb1 expression declines with aging and is specifically upregulated in lymphoid progenitors compared to HSCs [27]. Genetic deletion of Satb1 induced myeloid-biased differentiation and impaired T-cell lineage commitment, whereas its overexpression enhanced lymphoid differentiation [27]. Notably, whereas transient expression of Yamanaka factors (e.g., Oct4, Sox2, Klf4 and c-Myc) showed systemic rejuvenation effects and extended life span in mice [167], whether it can reverse HSC aging remains to be tested.

Small molecule and protein factor treatment

Florian et al. observed a marked increase in non-polarized cells among aged HSCs, which was attributed to elevated Rho-GTPase activity of Cdc42 during aging [29]. Casin, a small-molecule Cdc42 inhibitor, restored the proportion of polarized HSCs in aged mice and moderately enhanced their long-term reconstitution potential [30]. In vivo Casin treatment significantly extended the lifespan of aged mice and reduced systemic inflammatory cytokines [30]. Furthermore, ex vivo Casin treatment partially rejuvenated human aged HSCs, recapitulating functional restoration observed in murine models [31]. Aged HSCs exhibit MMP reduction and pronounced mitochondrial heterogeneity, with low-activity subpopulations displaying characteristic aging phenotypes [168]. Mito-Q treatment in aged mice partially restored MMP and augmented transcriptional activity in HSCs [12]. Competitive transplantation assays demonstrated Mito-Q-treated aging HSCs exhibited superior reconstitution capacity compared to untreated counterparts [12].

Protein factor supplementation is an alternative strategy for HSC rejuvenation. Guidi et al. observed age-associated reduction of Osteopontin (OPN) expression in BMSCs [32]. OPN treatment partially restored HSC aging phenotypes in vitro, and showed modest functional improvement after in vivo transplantation as compared to untreated aged HSCs [32]. IGF1 is another critical factor that shows age-dependent downregulation in the BM microenvironment [33]. Exogenous IGF1 stimulation reduced DNA damage, and improved cell polarization and mitochondrial function in middle-aged LT-HSCs, with selective enhancement of lymphoid-biased HSCs through increased cell division and numbers [33]. However, in vivo targeted delivery of IGF-1 to HSCs has not been achieved thus far.

Senolytic drugs and targeted removal of aged HSCs

Chang et al. found that oral administration of ABT263, an inhibitor of the anti-apoptotic proteins BCL-2 and BCL-xL, effectively cleared senescent HSCs, ameliorated irradiation-induced premature hematopoietic aging, and partially restored the regenerative ability of HSCs [34]. Wang et al. found detrimental effects of aged CD150^high HSCs on youthful CD150^low counterparts [35]. Transplantation experiments revealed that aged mice receiving CD150^low HSCs showed improved hematopoiesis and immune parameters, reduced epigenetic aging signatures in HSCs, as well as improved exercise capacity, muscular performance, and neurological functions as compared to CD150^high HSC recipients. Notably, these phenotypic improvements correlated with extended lifespan [35]. Similarly, Ross et al. defined two functionally distinct HSC subtypes based on CD150 expression levels: lympho-myeloid balanced HSCs (bal-HSCs, CD150^low) maintaining multilineage potential, and myeloid-biased HSCs (my-HSCs, CD150^high) exhibiting restricted differentiation [36]. Quantitative analysis revealed that my-HSCs expanded with aging as compared to bal-HSCs. Targeted depletion of my-HSCs in vivo using an anti-CD150 antibody effectively reversed this imbalance, thereby enhancing lymphopoiesis in aged mice [36]. These findings collectively implicate aged HSCs as a key contributor to organismal aging, and demonstrate that HSC sub-population rebalancing could serve as a systemic anti-aging strategy.

Heterochronic parabiosis

While young blood exhibits partial rejuvenation capacity for aged tissues including brain and muscle [169, 170], its effects on the aging hematopoietic system remain controversial. Ho et al. showed limited functional rejuvenation of aged HSCs in a heterochronic parabiosis model [160]. However, a multiomics study demonstrated that heterochronic parabiosis can significantly reverse the transcriptional landscape of aged HSCs [171]. Conventional myeloablative conditioning in HSC transplantation inevitably damages the BM niche. To circumvent this limitation, Kuribayashi et al. transplanted supraphysiological numbers of aged HSCs into unirradiated young recipients [172]. While aged HSCs successfully engrafted into young BM, they maintained persistent myeloid differentiation bias [172]. Transcriptomic and epigenetic analyses in these HSCs showed mild transcriptional but not DNA methylation changes, suggesting that the young BM microenvironment in vivo has limited rejuvenation effects on aged HSCs [172].

Ex vivo reconstitution of young BM niche

HSCs exhibit an accelerated tendency toward differentiation and impaired self-renewal upon leaving the BM niche. This is mainly because conventional culture conditions are incapable of mimicking the biochemical and biomechanical microenvironment of the native BM [173]. To address this, Nakahara et al. generated revitalized BMSCs by ectopic expression of 5 TFs, which significantly restored HSC niche factor expression and displayed a superior capacity to promote HSC expansion after 2D co-culture [174]. Donnelly et al. reported that a soft type-I collagen hydrogel maintains functional human CD34⁺ HSCs after co-culture with perivascular stromal cells [175]. ECs are also key niche components that regulate HSC function. Poulos et al. demonstrated that aged ECs impair the regenerative capacity of HSCs and induce myeloid lineage bias [37]. Transplantation of young ECs significantly restored hematopoietic reconstitution by aged HSCs and improved overall survival of lethally irradiated recipients, although it failed to reverse the intrinsic myeloid lineage predisposition [37].

Recently, we demonstrated that HSCs can be robustly rejuvenated by 3D co-culture with BMSCs in a soft gelatin methacryloyl (GelMA) hydrogel [19, 176]. We showed that BMSCs dramatically down-regulate HSC niche factors, such as Scf and Cxcl12, after 2D culture on stiff plastic dish, which could be significantly restored upon 3D culture in soft GelMA hydrogel [19]. Mechanistic studies revealed that Yap/Taz activation promotes BMSC proliferation and inhibits niche factor expression. More importantly, 3D co-culture significantly increased lymphopoietic differentiation, reversed HSC aging markers, and completely restored the long-term multilineage reconstitution capacity of middle-aged or aged HSCs [19]. Finally, by in situ measurement of the BM stiffness using AFM, we demonstrated that BM stiffening is a novel hallmark of hematopoietic aging. Therefore, biomechanical factors like the matrix stiffness critically regulate HSC aging by BMSCs, which could be targeted to promote HSC rejuvenation [19].

Conclusions and perspectives

Inspection of the HSC life cycle reveals an interesting phenomenon: lymphopoietic potential reflects the maturity and robustness of the hematopoietic and immune system, which peaks in adulthood but is largely dispensable during early development and aging. As HSCs age, they undergo intrinsic functional decline, characterized by accumulated DNA damage, telomere shortening, epigenetic drift, dysregulated proteostasis, and mitochondrial dysfunction. Meanwhile, their BM microenvironment also exhibits degenerative changes such as inflammatory factor accumulation, vasculature disruption, and BM stiffening. These alterations collectively drive diminished HSC self-renewal capacity, myeloid-biased differentiation, compromised long-term multilineage reconstitution and CH, ultimately leading to anemia, immune aging, and increased risks of hematologic malignancies. Notably, a bidirectional feedback cycle exists between HSC aging and microenvironmental deterioration: microenvironmental dysfunction accelerates HSC aging, while aberrant clonal expansion of HSCs further disrupts microenvironmental homeostasis, thereby exacerbating age-related hematopoietic pathologies.

It is worth noting that most HSC aging studies have relied on cells isolated from limb bones (femur and/or tibia). However, emerging evidence reveals that HSCs from other skeletal sites may exhibit fundamentally different aging dynamics. The skull BM, connected to the meninges via specialized vascular channels, harbors HSCs that directly contribute to immune surveillance in the central nervous system [177, 178]. Remarkably, a recent study demonstrated that skull BM undergoes lifelong expansion driven by VEGFA-dependent vascular growth, and is largely protected against major hallmarks of BM aging—including adipocyte accumulation, pro-inflammatory cytokine upregulation, and vascular integrity loss—that are prominent in femoral BM [179]. Moreover, the systemic hematopoietic contribution from skull BM increases with age, and skull BM exhibits significantly less myeloid bias than femur in geriatric mice. These findings suggest that the anatomical origin of HSCs may profoundly influence conclusions about aging mechanisms and rejuvenation efficacy. Future studies should systematically compare HSC properties across different skeletal compartments, which could reveal site-specific regulatory mechanisms currently masked by exclusive focus on limb bones, in order to identify novel therapeutic targets for HSC rejuvenation.

Emerging new technologies are providing powerful tools to advance HSC aging studies. Spatial transcriptomics and proteomics enable in situ detection of the spatiotemporal dynamics of HSC-niche interaction during aging [180, 181]. AFM enables accurate assessment of age-related alterations in BM niche stiffness [19]. Lipid nanoparticle (LNP) enables targeted delivery of rejuvenating factors to HSCs or their niche for in vivo intervention [182]; CAR-T cell technology enables targeted depletion of aged HSCs to create a youthful BM niche [183]. EPI-Clone technology enables high throughput reconstruction of clonal differentiation trajectories, which could help elucidate HSC clonal diversity during aging and rejuvenation [184]. Novel epigenetic editing tools enable correction of age-related DNA methylation or histone modification aberrations, thereby restoring HSC stemness and differentiation balance. Finally, artificial intelligence empowers aging clock studies that will inevitably accelerate the discovery of novel aging biomarkers, and guide rational design of cutting-edge rejuvenation strategies as multi-modal datasets continue to expand. By developing these new technologies, we expect to witness a broader and more successful application of stem cell therapy in anti-aging industries.

Authors’ contributions

R.Y. and Y.Z. conceived and wrote the manuscript. Y.X., W.S., D.C. and D.G. helped revise the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by grants from the Shenzhen Medical Research Fund (B2402003), National Key R&D Program of China (2022YFA1103200 and 2021YFA1100900), National Natural Science Foundation of China (32425027, 32330050, 82361148131, 82302706), Shanghai Municipal Science and Technology Commission (23XD1423900), Shanghai Pilot Program for Basic Research, Peak Disciplines (Type IV) of Institutions of Higher Learning in Shanghai, Fundamental Research Funds for the Central Universities (22120250374), and Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China (JYB2025XDXM508). R.Y. is a SANS exploration scholar.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.