Stem cell senescence as a regenerative bottleneck in age-related skeletal degeneration: Mechanistic crosstalk and translational strategies

Abstract

Skeletal degeneration is a major cause of pain, disability, fragility fracture, and loss of independence in aging populations. Although osteoporosis, osteoarthritis, and intervertebral disc degeneration are usually managed as anatomically separate disorders, they share a common regenerative failure: endogenous stem and progenitor cells lose renewal capacity, acquire inflammatory secretory programs, and become less able to rebuild damaged matrix. This narrative review synthesizes evidence on senescence in bone marrow mesenchymal stem/stromal cells, skeletal stem cells, hematopoietic stem cells, cartilage progenitor cells, synovial mesenchymal stromal cells, and disc-derived progenitor populations. Priority is given to primary mechanistic, omics, translational, and early clinical studies that connect senescence markers with functional impairment or intervention response. We discuss telomere-associated DNA damage, mitochondrial and oxidative stress, nutrient- and mechanosensing pathways, epigenetic remodeling, dysregulated autophagy and mitophagy, and the senescence-associated secretory phenotype as interlocking mechanisms that reshape skeletal niches. Disease-specific sections then examine how senescent progenitor compartments drive osteoporosis, osteoarthritis, and intervertebral disc degeneration through impaired osteogenesis or chondrogenesis, matrix catabolism, inflammatory vesicle signaling, and maladaptive mechanical feedback. Finally, we compare senolytics, senomorphics, metabolic and mitochondrial regulators, cell rejuvenation, extracellular vesicle engineering, biomaterials, and gene-based delivery platforms from a regenerative medicine perspective. Stem/progenitor cell senescence should be treated not simply as a biomarker of aging, but as a compartment-specific therapeutic bottleneck. Translation will require rigorous cell identity, multi-parameter senescence assays, disease-stage stratification, and therapies that restore regenerative competence without suppressing physiological repair.

Highlights

• •Stem cell senescence links skeletal aging to regenerative failure.
• •SASP connects inflammation, mechanics, and matrix degradation.
• •Senolytics, EVs, biomaterials, and gene delivery are compared.
• •Disease-stage and compartment matching are key for translation.
• •Multi-parameter senescence assays are needed before clinical use.

Abbreviations

BMSCs

bone marrow mesenchymal stem/stromal cells

EVs

extracellular vesicles

HSCs

hematopoietic stem cells

IDD

intervertebral disc degeneration

IVD

intervertebral disc

MSCs

mesenchymal stem/stromal cells

NP

nucleus pulposus

OA

osteoarthritis

ROS

reactive oxygen species

SASP

senescence-associated secretory phenotype

1. Introduction

The clinical burden of skeletal degeneration is expanding as populations age and as obesity, inactivity, endocrine disorders, and multimorbidity increase the lifetime exposure of bone, cartilage, and intervertebral disc tissues to metabolic and mechanical stress. Global disease-burden analyses show that musculoskeletal disorders and low back pain remain leading contributors to years lived with disability, while osteoarthritis and fragility fractures produce large and growing demands on health systems [[1], [2], [3], [4]]. These disorders are often treated as anatomically separate conditions, but their biological trajectories converge on a common failure of tissue maintenance: bone formation becomes uncoupled from resorption, articular cartilage loses matrix integrity and repair competence, and the intervertebral disc shifts toward catabolic remodeling and structural collapse.

Skeletal tissue homeostasis depends on coordinated activity among multiple stem and progenitor cell populations. Bone marrow mesenchymal stem/stromal cells (BMSCs) generate osteoblast-lineage cells, adipocytes, chondrocyte-like cells, and stromal niche components; skeletal stem cells and periosteal progenitors contribute to growth and fracture repair; hematopoietic stem cells (HSCs) shape immune output in the marrow niche; cartilage progenitor cells and synovial mesenchymal stromal cells influence articular repair; and disc-derived progenitor populations support nucleus pulposus and annulus fibrosus homeostasis [[5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]]. These populations are not static. Their function is shaped by chronological age, inflammatory exposure, endocrine state, oxygen tension, extracellular matrix quality, and mechanical loading.

Cellular senescence is a durable stress-response state characterized by cell-cycle arrest, altered chromatin organization, resistance to apoptosis, metabolic rewiring, and secretion of inflammatory and matrix-modifying factors [[17], [18], [19], [20]]. In stem and progenitor cells, senescence is especially consequential because it reduces regenerative reserve and can transmit dysfunction to neighboring cells. Transgenic and pharmacologic studies show that senescent-cell accumulation contributes to age-related tissue decline and that selective clearance can improve tissue function in several models [[21], [22], [23], [24], [25], [26], [27]]. In bone, cartilage, and disc tissues, the evidence is increasingly mechanistic: senescent progenitor compartments display reduced clonogenicity, impaired osteogenic or chondrogenic differentiation, adipogenic or fibro-inflammatory lineage skewing, abnormal extracellular vesicle cargo, and SASP-driven effects on osteoclasts, immune cells, chondrocytes, and matrix turnover [[28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40]].

This review is framed for regenerative therapy. Its central premise is that stem/progenitor cell senescence is not only a marker of tissue aging but also a modifiable bottleneck that limits cell therapy, extracellular vesicle therapy, biomaterial-assisted repair, and endogenous tissue regeneration. The review therefore integrates molecular mechanisms, disease-specific evidence in osteoporosis, osteoarthritis, and intervertebral disc degeneration, reciprocal microenvironmental crosstalk, and intervention strategies. Its added value is the compartment-resolved focus on stem/progenitor populations, which helps distinguish causal senescence mechanisms from broader age-associated tissue damage and provides a practical framework for matching therapies to disease stage, tissue niche, and regenerative objective.

2. Review strategy and evidence prioritization

This article was prepared as a narrative review with evidence prioritization rather than as a systematic review. The literature was screened around the concepts of stem/progenitor cell senescence, BMSCs and skeletal stem cells, osteoporosis, osteoarthritis, intervertebral disc degeneration, senolytics, senomorphics, extracellular vesicles, biomaterials, gene delivery, and regenerative medicine. Priority was given to recent primary experimental, single-cell or multi-omics, translational, and clinical studies when they directly supported a mechanistic or therapeutic statement.

Evidence was interpreted in four tiers: direct intervention studies showing senescent-cell clearance or pathway rescue; primary experimental and omics studies linking senescence markers with functional stem/progenitor impairment; translational or early clinical studies of cell-, vesicle-, biomaterial-, or senescence-targeted therapies; and foundational reviews or consensus statements used mainly to define terminology or standards. Particular attention was given to whether a study identified the relevant cell population, measured more than one senescence feature, and linked cellular findings to tissue-level outcomes. Because no formal systematic-review protocol or quantitative meta-analysis was used, therapeutic efficacy is interpreted conservatively and framed according to evidence strength, disease compartment, and translational readiness.Across the disease-specific sections, evidence type is specified where relevant as cultured-cell, animal/in vivo, omics or tissue-profiling, translational, or early clinical evidence to help separate causal inference from associative observations.

3. Senescent stem/progenitor compartments in skeletal tissues

The term BMSC is often used broadly, yet bone marrow stromal preparations contain heterogeneous subsets with variable self-renewal, osteogenic, adipogenic, immunomodulatory, and niche-supporting capacities. Functional transplantation, lineage tracing, and single-cell analysis have identified self-renewing osteoprogenitors, mouse skeletal stem cells, human skeletal stem cells, growth-plate skeletal stem cells, periosteal stem cells, and diverse marrow stromal populations [[6], [7], [8], [9], [10], [11], [12], [13], [14], [15]]. These studies demonstrate that skeletal stem/progenitor identity is context-dependent and spatially organized; senescence in one compartment should therefore not be generalized automatically to all skeletal tissues.

Senescence is also operationally complex. No single marker is sufficient. Increased p16INK4a, p21CIP1, p53 activation, senescence-associated beta-galactosidase staining, DNA-damage foci, telomere dysfunction-induced foci, lamin B1 loss, mitochondrial depolarization, altered NAD + metabolism, and SASP cytokines each indicate aspects of a senescent state, but their specificity varies by cell type and stimulus [[21], [22], [23], [24], [25], [26], [27]]. In stem/progenitor cells, the most meaningful definition combines molecular markers with functional assays, including colony formation, serial passaging, lineage differentiation, transplantation or repair competence, and paracrine effects on neighboring cells.

Recent geroscience frameworks further caution against treating senescence as a single uniform state [[41], [42], [43], [44]]. In skeletal tissues, early senescence may protect against malignant transformation or excessive proliferation after injury, whereas persistent senescence in aged or chronically inflamed tissues becomes maladaptive. This maladaptation appears as lower osteoblastogenesis, greater marrow adipogenesis, impaired cartilage matrix synthesis, reduced disc-cell survival, increased osteoclast-supportive signaling, and chronic innate immune activation. The biological question is therefore not simply whether senescent cells are present, but which stem/progenitor subsets become senescent, when this occurs during disease progression, and how the resulting secretome and lineage bias alter the local niche (Table 1).

Table 1.

Stem/progenitor cell populations relevant to skeletal senescence.

Cell population	Main skeletal role	Senescence-related disease effect	Representative evidence
BMSCs/marrow stromal cells	Osteogenic, adipogenic, chondrogenic and niche-supporting functions	Reduced clonogenicity, osteogenic decline, adipogenic skewing, inflammatory EV cargo	[16,[24], [25], [26], [27], [28], [29], [30], [31],[37], [38], [39], [40], [41], [42], [43], [44], [45], [46]]
Skeletal stem cells/periosteal stem cells	Bone growth, fracture repair, osteochondral regeneration	Inflammatory degenerative niche, impaired osteochondrogenesis, pro-osteoclastic signals	[[7], [8], [9], [10], [11], [12], [13], [14], [15],21]
HSCs and immune-biased progenitors	Immune output and marrow niche regulation	Myeloid skewing and inflammatory reinforcement of skeletal degeneration	[17,18,33]
Cartilage progenitor and synovial stromal cells	Cartilage repair and joint immunomodulation	Loss of chondrogenic competence, SASP-mediated catabolism and synovitis	[[47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65]]
Disc-derived stem/progenitor cells	Nucleus pulposus and annulus fibrosus maintenance	Reduced survival in acidic/hypoxic niche and catabolic inflammatory signaling	[[66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77]]

Note: BMSCs, bone marrow mesenchymal stem/stromal cells; EV, extracellular vesicle; HSCs, hematopoietic stem cells; SASP, senescence-associated secretory phenotype.

A useful therapeutic distinction is between cell-autonomous and non-cell-autonomous senescence. Cell-autonomous senescence arises when intrinsic damage, such as telomere dysfunction, mitochondrial failure, DNA damage, or epigenetic drift, directly reduces stem/progenitor fitness. Non-cell-autonomous senescence arises when the surrounding niche imposes inflammatory, mechanical, or metabolic pressure on otherwise competent cells. In practice, both mechanisms coexist. A low-senescence MSC product may fail in a severely inflamed joint unless synovitis and matrix catabolism are controlled, whereas senomorphic suppression of inflammation may not rebuild tissue if the endogenous progenitor pool is already exhausted.

4. Core molecular mechanisms of stem/progenitor cell senescence

Multiple upstream insults converge on a limited set of senescence outputs in skeletal stem/progenitor cells. Telomere erosion and DNA damage activate ATM/ATR-p53-p21 and p16INK4a-RB checkpoints, thereby restricting proliferation and promoting stable arrest. In BMSCs, telomere dysfunction can be coupled to mitochondrial metabolic failure and reduced osteogenic capacity, suggesting that replicative aging and bioenergetic decline reinforce one another [28]. Similar logic applies to aged skeletal stem cells, in which reduced osteochondrogenic output occurs alongside inflammatory remodeling of the niche [15]. These data support telomere-associated stress as a plausible contributor to skeletal aging, but also show that telomere attrition rarely acts alone.

Oxidative stress and mitochondrial dysfunction form a second mechanistic axis. Excess reactive oxygen species damage DNA, proteins, and lipids, while impaired mitochondrial quality control reduces ATP availability and amplifies stress signaling. BMSCs exposed to aged bone-matrix extracellular vesicles show altered lineage allocation, and mitochondrial transfer can enhance BMSC proliferation, migration, osteogenic differentiation, and bone-defect repair in experimental systems [30,36]. In HSCs, mitochondrial unfolded-protein responses are activated during exit from quiescence, illustrating that mitochondrial stress can influence stem-cell state transitions [37]. Studies of cadmium exposure, SIRT3 sulfhydration, and autophagy-related regulators further show that mitochondrial and redox pathways can directly modulate osteogenic failure and senescence-associated bone loss [[31], [32], [33], [34], [35]].

Nutrient-, stress-, and mechanosensing pathways provide a third layer of regulation. mTOR integrates amino-acid, growth-factor, and energy signals; excessive or sustained mTOR activation can promote cellular hypertrophy, SASP production, and senescence-like growth arrest. Wnt/beta-catenin signaling maintains osteogenic and chondrogenic programs, but abnormal activation or suppression can produce lineage imbalance. NF-kappaB links inflammatory signaling to SASP gene expression, while focal adhesion kinase, integrin-linked kinase, MAPK cascades, and related mechanotransduction pathways translate surface topography and mechanical cues into transcriptional outputs [[38], [39], [40]]. These pathways are best understood as a network rather than as isolated linear mechanisms.

Epigenetic remodeling and RNA-mediated regulation explain why senescence phenotypes may persist after the original stress is removed. DNA methylation drift, histone modification changes, chromatin accessibility shifts, and non-coding RNAs can lock cells into lineage-biased or inflammatory states. Experimental work has linked BMSC senescence and osteoporotic phenotypes to miR-203-3p/Pbk signaling, NAP1L2-associated chromatin regulation, and other non-coding RNA networks [32,33]. Multi-omics analyses of BMSCs in osteoporosis have also identified disease-associated changes in differentiation-related regulatory circuits [45,46]. The therapeutic appeal of this axis is reversibility, but specificity remains difficult because a regulatory RNA or chromatin factor may have different effects across osteogenic, adipogenic, chondrogenic, and immune-supportive subsets.

The SASP is the main mechanism by which senescent stem/progenitor cells influence tissue-level disease. SASP components include IL-6, IL-8, TNF-alpha, CCL2, CXCL chemokines, matrix metalloproteinases, growth factors, prostaglandins, and extracellular vesicle cargo. In bone-derived extracellular vesicles, age-associated miRNAs can suppress BMSC proliferation and induce senescence-like behavior [29,30]. In articular and disc tissues, SASP factors and vesicle cargo can promote macrophage activation, chondrocyte catabolism, matrix degradation, osteoclastogenesis, nociceptive inflammatory signaling, and secondary senescence [[47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69]]. Because SASP composition is stimulus- and tissue-specific, future studies need to define which secretory modules are destructive, which are compensatory, and which can be reshaped without impairing repair (Fig. 1).

Fig. 1.

Core molecular mechanisms of stem cell senescence in skeletal degeneration. The figure summarizes telomere attrition and telomerase dysfunction, oxidative stress, dysregulated signaling, epigenetic alterations, mitochondrial dysfunction, defective mitophagy, and SASP activation as convergent mechanisms leading to stem/progenitor cell senescence, cell-cycle arrest, lineage imbalance, and functional loss. Cell-type context: telomere- and DNA-damage-related arrest has been described mainly in BMSCs and skeletal stem/progenitor cells; mitochondrial/redox dysfunction and defective autophagy or mitophagy have been reported in BMSCs, HSCs during state transitions, adipose-derived MSCs, and nucleus-pulposus or disc-derived progenitor systems; mTOR, Wnt/beta-catenin, FAK/MAPK, and ILK/beta-catenin mechanosensing pathways have been investigated primarily in BMSCs and related mesenchymal/skeletal progenitors; and SASP or extracellular-vesicle signaling has been shown in BMSCs, cartilage progenitor or synovial stromal compartments, and nucleus-pulposus/disc-derived cells.

5. Disease-specific evidence

5.1. Osteoporosis

Osteoporosis is characterized by reduced bone mass, microarchitectural deterioration, and increased fracture susceptibility. The evidence linking stem/progenitor cell senescence to osteoporosis is comparatively strong, but its evidentiary weight differs by study type. Human donor studies and multi-omics analyses mainly support an association between aged or osteoporotic BMSCs and lower clonogenicity, weaker osteogenic differentiation, adipogenic skewing, mitochondrial dysfunction, and increased senescence-marker expression [16,28,45,46,[70], [71], [72], [73], [74]]. Animal and intervention studies provide stronger causal evidence: senescent-cell targeting attenuates age-related bone loss in mice, and aged skeletal stem-cell lineages can generate inflammatory and osteoclast-supportive niches that impair regeneration [15,25]. Thus, the most reliable conclusion is that BMSC and skeletal stem-cell senescence is not merely a marker of low bone mass; it is a biologically active contributor to failed osteoblast-lineage renewal, although the relative contribution of each progenitor subset remains model- and compartment-dependent.

Female sex hormones are an essential clinical modifier of this interpretation. In postmenopausal or estrogen-deficient osteoporosis, loss of estrogenic signaling increases bone resorption, inflammatory tone, oxidative stress, and marrow adiposity, thereby creating a systemic niche that can accelerate BMSC dysfunction and senescence-associated loss of osteogenic commitment. This endocrine layer is important because osteoporosis in women should not be read as a purely chronological-aging phenotype: estrogen deficiency can amplify the same regenerative bottleneck through osteoclast activation, stromal inflammatory remodeling, and adipogenic lineage bias. For regenerative therapy, menopausal status, sex-hormone exposure, and antiresorptive or anabolic treatment history should therefore be treated as biological context rather than background clinical descriptors [74,77].

Mechanistic osteoporosis studies identify candidate regulators that are directly relevant to regenerative therapy, but they should be interpreted as convergent nodes rather than isolated causal pathways. SIRT3 sulfhydration stabilizes heterochromatin and mitochondrial homeostasis in BMSCs, miR-203-3p promotes BMSC senescence through Pbk downregulation, NAP1L2 drives mesenchymal stem-cell senescence and impaired osteogenic differentiation, and OPTN-dependent autophagy protects against osteoporotic changes by regulating lipid-handling pathways [[31], [32], [33], [34]]. Aged bone matrix-derived extracellular vesicles can transfer signals that favor BMSC adipogenesis over osteoblastogenesis, offering a paracrine explanation for the coexistence of skeletal demineralization and marrow adiposity [30]. The value of these cultured-cell, animal, and omics studies is that they converge on a consistent regenerative phenotype: mitochondrial/redox injury, chromatin dysregulation, autophagy failure, and EV-mediated niche signaling reduce osteoblast-lineage renewal while increasing adipogenic or inflammatory outputs.

Therapeutic evidence remains largely preclinical but informative when evaluated through this evidence-strength lens. Melatonin has been reported to enhance osteoblastogenesis of senescent BMSCs; icariin, a prenylated flavonol glycoside derived from Epimedium species, has been investigated for its ability to influence osteogenic/adipogenic allocation; and BMSC-derived or engineered extracellular vesicles have been tested for bone-regenerative effects [[74], [75], [76], [77]]. Senolytic strategies improve the bone-forming potential of aged marrow stromal cells in experimental systems, and bone-targeted delivery increases skeletal selectivity while reducing the conceptual risk of systemic exposure [78,79]. Parallel work suggests that restoring autophagy-lysosomal programs or mechanically rejuvenating senescent BMSCs can reopen osteogenic potential through TFEB- and chromatin-linked mechanisms [80,81]. However, these approaches are not yet equivalent to established clinical osteoporosis therapies. Their translational promise rests on whether they can improve fracture resistance, marrow adiposity, bone turnover, cellular senescence biomarkers, and repair quality in older patients with endocrine changes and comorbidities, rather than improving mineralization endpoints alone.

5.2. Osteoarthritis

Osteoarthritis involves articular cartilage degeneration, synovial inflammation, subchondral bone remodeling, osteophyte formation, and pain. Stem/progenitor cell senescence is relevant because cartilage has limited intrinsic repair capacity and because synovial and subchondral compartments actively communicate with cartilage. The strongest causal evidence comes from senescent-cell clearance in post-traumatic osteoarthritis models, where local elimination of senescent cells reduced cartilage degeneration and promoted a more regenerative environment [26]. More recent multi-omic work identified senescent stem cell populations in subchondral bone as contributors to osteoarthritis progression and potential therapeutic targets [47]. This section therefore distinguishes animal proof-of-concept data, single-cell or spatial profiling of diseased tissues, and early clinical MSC studies rather than treating all cited evidence as equivalent.

Several stem-cell compartments may participate. Cartilage progenitor cells and synovial mesenchymal stromal cells may lose chondrogenic competence under inflammatory and mechanical stress, while BMSCs in subchondral bone can acquire osteoclast-supportive or fibrotic programs. Adipose-derived mesenchymal stromal cells are clinically accessible, but their therapeutic potential depends on donor age, culture conditions, senescence burden, and secretome quality. Experimental studies show that curcumin can modulate adipose-derived MSC senescence through FoxO3/autophagy signaling and that stem-cell-derived extracellular vesicles can reduce cartilage destruction, modulate macrophage polarization, and support matrix homeostasis in animal models [[48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59]]. Single-cell and synovial analyses further identify senescence-associated stromal and immune niches in osteoarthritic joints [60].

Clinical evidence for intra-articular MSC therapy in osteoarthritis is promising but still inconsistent. Early proof-of-concept and randomized studies suggest potential improvements in symptoms, cartilage quality, or structural biomarkers in selected patients, but effect sizes vary and product heterogeneity remains substantial [[82], [83], [84], [85]]. This heterogeneity should not be interpreted simply as failure of the stem-cell concept. Instead, it indicates that therapeutic cells and vesicles must be characterized for senescence markers, secretome composition, passage number, donor variability, potency, and joint-stage compatibility. A cell product enriched in senescent or inflammation-primed cells could plausibly worsen the microenvironment, whereas a low-senescence, chondrogenic, immunomodulatory product may be beneficial.

The osteoarthritis field is shifting toward spatial and single-cell resolution. Multi-omic and compartment-level studies have identified senescent stromal and stem-like populations in subchondral bone, synovium, and bone marrow lesions, while recent mechanistic work suggests that senescent BMSCs can aggravate subchondral sclerosis and cartilage degeneration through SASP-mediated remodeling [[86], [87], [88], [89]]. These data are valuable because they connect senescence to anatomically defined disease niches rather than treating osteoarthritis as a purely cartilage-intrinsic disorder.

5.3. Intervertebral disc degeneration

Intervertebral disc degeneration is characterized by nucleus pulposus dehydration, extracellular matrix breakdown, annulus fibrosus damage, endplate sclerosis, inflammatory signaling, and pain-related neural and vascular ingrowth. The disc niche is hypoxic, acidic, hyperosmolar, and nutrient-limited, which creates a hostile environment for both resident progenitor cells and transplanted cells. Nucleus pulposus mesenchymal stem/progenitor cells display greater adaptation to disc-like acidity than adipose-derived MSCs in experimental comparisons, but degenerative stress still reduces their viability, proliferation, and matrix-anabolic activity [61,62]. The evidence base in this section is mainly composed of cultured-cell experiments, exosome or biomaterial studies, animal/ex vivo disc models, single-cell or proteomic tissue profiling, and early translational cell-therapy reports.

Primary studies indicate that inflammatory vesicle signaling, senescent nucleus pulposus proteomic profiles, and progenitor-cell rejuvenation pathways are involved in disc degeneration [[63], [64], [65], [66], [67], [68], [69]]. M1 macrophage-derived exosomes can promote nucleus pulposus cell senescence through LCN2/NF-kappaB signaling, linking immune activation to disc-cell aging [64]. Proteomics of senescent nucleus pulposus tissue provides molecular candidates that may explain matrix catabolism and stress adaptation [65]. Disc or nucleus-pulposus cell-derived extracellular vesicles have also shown therapeutic potential in experimental models, suggesting that cell-free approaches may better tolerate the disc microenvironment than direct cell transplantation in some settings [67,68].

The major translational challenge is delivery. Injected cells must survive low glucose, low oxygen, acidity, high osmolarity, and compressive loading. Biomaterials, hydrogels, preconditioning, and extracellular-vesicle approaches are therefore central to disc regeneration strategies [[90], [91], [92], [93]]. Recent studies emphasize intercellular communication: degenerative nucleus pulposus cell-derived vesicles, macrophage-derived exosomes, and MSC-mediated oxidative-stress and autophagy pathways can influence nucleus pulposus senescence, cartilage endplate injury, macrophage polarization, and matrix breakdown [[94], [95], [96]]. Biomaterial strategies that target senescent cells or provide protective delivery environments may therefore be most useful when they combine anti-senescent activity with niche reconstruction rather than functioning only as passive carriers [97].

6. Reciprocal crosstalk between senescent stem cells and the skeletal microenvironment

The skeletal microenvironment is not a passive background for stem-cell aging. It determines whether a stem/progenitor cell remains quiescent, differentiates, repairs tissue, becomes exhausted, or enters senescence. Conversely, senescent stem/progenitor cells reshape the niche through SASP factors, extracellular vesicles, altered matrix deposition, and defective lineage output. This bidirectional relationship explains why skeletal degeneration often progresses after the initiating insult has passed.

Inflammation is a central amplifier. TNF-alpha, IL-1beta, IL-6, prostaglandins, and macrophage-derived factors can suppress osteogenesis, enhance osteoclastogenesis, and induce senescence-associated signaling in BMSCs, chondrocytes, and disc cells. FGF2-mediated rescue of TNF-alpha-induced osteogenic damage in human BMSCs indicates that inflammatory injury can be partially reversible when mitochondrial oxidative phosphorylation is preserved [93]. In osteoarthritis and disc degeneration, macrophage polarization and vesicle cargo transmit inflammatory memory across compartments [[51], [52], [53], [54], [55],64,[94], [95], [96]]. HSC aging may further skew immune output toward myeloid and pro-inflammatory phenotypes, strengthening the link between marrow aging and skeletal disease progression [21,22].

Extracellular matrix remodeling is another major regulator. Matrix stiffness, collagen organization, fibronectin network integrity, proteoglycan loss, and decellularized matrix composition influence stem-cell adhesion, mechanotransduction, and differentiation. Decellularized matrix studies show that matrix source and quality can modify cartilage-regenerative behavior, while senescent-cell SASP proteases can accelerate matrix breakdown [98,99]. This creates a feed-forward loop: matrix degradation distorts stem-cell signaling, and senescent stem/progenitor cells produce factors that further degrade matrix. In osteoporosis this loop appears as weakened osteogenic support and marrow adiposity; in osteoarthritis it appears as cartilage loss and subchondral remodeling; and in disc degeneration it appears as aggrecan loss, collagen disorganization, and mechanical instability.

Mechanical stress provides a third layer of crosstalk. Physiological loading supports bone and cartilage homeostasis, whereas overload, abnormal shear, immobilization, and matrix stiffening can drive maladaptive mechanotransduction. BMSCs respond to surface topography through FAK/MAPK and ILK/beta-catenin pathways, and low-intensity pulsed ultrasound can influence osteogenic commitment under defined scaffold conditions [38,[100], [101], [102]]. At the same time, senescent stem/progenitor cells may lose normal mechanosensitivity, weakening their ability to adapt to loading. Mechanically informed biomaterials and loading protocols should therefore be designed alongside senescence biomarkers rather than treated as separate variables.

The immune component of the niche determines whether senescent cells are cleared or retained. In young tissues, senescent cells can be recognized and removed by macrophages, natural killer cells, and other immune cells. In aged or chronically inflamed skeletal tissues, immune clearance may become inefficient while inflammatory signals remain high. Matrix fragments, advanced glycation end products, and mineralized debris can further activate innate immune receptors. The microenvironmental loop therefore has at least three nodes: stem/progenitor senescence, inflammatory immune activation, and matrix/mechanical damage. Breaking only one node may provide partial benefit, but durable disease modification will likely require coordinated intervention (Fig. 2). Experimental tools such as exosome-regeneration systems, immune-orchestration models, synovium-derived stem-cell loading systems, tendon-bone differentiation models, and organ-on-chip platforms should be linked more directly to senescence readouts [[103], [104], [105], [106], [107], [108], [109]].

Fig. 2.

Disease-level consequences and microenvironmental feedback loops of senescent stem/progenitor cells. The figure illustrates how senescent BMSCs, cartilage progenitor cells, and disc-derived stem/progenitor cells contribute to osteoporosis, osteoarthritis, and intervertebral disc degeneration through chronic inflammation, extracellular matrix remodeling, and mechanical-stress dysregulation.

7. Therapeutic strategies targeting stem/progenitor cell senescence

Therapeutic development can be divided into four overlapping approaches: preventing premature senescence, suppressing harmful senescence signaling, eliminating deleterious senescent cells, and rebuilding a supportive niche. Senolytics such as BCL-2 family inhibitors and related agents exploit survival dependencies of senescent cells, whereas senomorphic agents aim to suppress SASP output without killing cells [[105], [106], [107], [108],[110], [111], [112], [113], [114]]. In skeletal disease, the most direct preclinical evidence comes from age-related bone loss and osteoarthritis models in which senescent-cell targeting improved tissue outcomes [25,26]. Nevertheless, senolysis must be used carefully because transient senescence can support wound healing, and indiscriminate elimination may impair repair or produce off-target toxicity.

Metabolic and mitochondrial regulators represent a complementary strategy. SIRT3 stabilization, melatonin, icariin, autophagy modulators, and interventions that improve mitochondrial quality can reduce oxidative damage and restore osteogenic or chondrogenic performance in selected models [31,34,35,74,75,80,81]. These agents may be especially useful when senescence is not fully established or when the goal is to improve resilience of endogenous stem/progenitor cells. Future studies should include parallel readouts for cell-cycle arrest, SASP composition, mitochondrial function, lineage output, and in vivo tissue repair rather than relying only on endpoint mineralization or cartilage staining.

Cell-based therapy has the advantage of delivering viable cells with regenerative and immunomodulatory capacity, but it also carries the risk of culture-induced senescence, donor variability, low persistence, and inconsistent potency. Clinical osteoarthritis studies illustrate both promise and uncertainty [[82], [83], [84], [85]]. Product release criteria should include senescence burden, passage number, proliferation history, differentiation potential, secretome profile, sterility, chromosomal stability, and immunomodulatory potency. Preconditioning strategies such as hypoxia, growth-factor priming, controlled mechanical stimulation, and mitochondrial enhancement may improve cell quality, but they require standardized protocols and safety testing. In disc degeneration, survival within the acidic and nutrient-poor niche is a particularly stringent barrier [61,62,[90], [91], [92], [93]].

Extracellular vesicle therapy may overcome some limitations of whole-cell transplantation. MSC-derived and tissue-derived vesicles can transfer regulatory RNAs, proteins, and lipids; in skeletal models they have been reported to reduce inflammation, improve matrix balance, and promote bone, cartilage, or disc repair [[50], [51], [52], [53], [54], [55], [56], [57], [58], [59],68,76,77]. Engineering strategies can enhance targeting, cargo loading, and potency, as shown by cartilage-targeted vesicles and METTL3-related exosome platforms [51,77,111]. However, EV therapy is not automatically safer or more reproducible than cell therapy. Vesicle isolation method, donor source, passage number, culture medium, storage, dose, biodistribution, and potency assays strongly influence results. Regulatory translation will require manufacturing standards that define identity, purity, potency, and mechanism of action [[112], [113], [114]].

Gene- and biomaterial-based approaches are emerging but remain early in translation. Gene modulation can target p16INK4a/p21, SIRT pathways, miRNAs, m6A regulators, autophagy genes, or lineage-specifying transcription factors; biomaterials can provide mechanical support, sustained release, and cell-protective microenvironments [77,[90], [91], [92], [93],97,115]. The most realistic near-term applications may not be permanent gene editing but transient RNA, vesicle, or biomaterial delivery designed to restore a regenerative window after injury or early degeneration. For clinical use, the field needs compartment-specific targeting. Osteoporotic marrow, osteoarthritic synovium, subchondral bone, articular cartilage, and the intervertebral disc differ markedly in vascularity, immune access, mechanical loading, and turnover. A single systemic anti-senescence therapy is unlikely to be optimal for all (Table 2).

Table 2.

Therapeutic strategies targeting stem/progenitor cell senescence in skeletal degeneration.

Therapeutic class	Mechanistic rationale	Evidence status	Key limitations
Senolytics	Remove cells dependent on anti-apoptotic survival networks	Strong proof-of-concept in aging, bone loss and OA models	Cell-type specificity, dosing, off-target toxicity, impaired repair risk
Senomorphics/anti-inflammatory agents	Suppress SASP, NF-kappaB, oxidative and inflammatory signaling	Preclinical evidence across BMSC, OA and disc models	May not restore exhausted progenitor pools
Metabolic and mitochondrial regulators	Improve redox balance, mitophagy, NAD+/sirtuin and ATP homeostasis	Preclinical evidence for BMSC function and bone repair	Mechanism often indirect; biomarkers need standardization
Cell rejuvenation and transplantation	Deliver low-senescence cells or restore endogenous cell potency	Early clinical and preclinical studies in OA and IVD disease	Donor variability, passage effects, persistence, potency release criteria
Extracellular vesicle therapy	Transfer regulatory RNAs/proteins with lower engraftment demands	Rapidly expanding preclinical evidence in bone, cartilage and disc models	Batch heterogeneity, targeting, dose, biodistribution and potency assays
Gene and biomaterial platforms	Modulate upstream regulators and reconstruct supportive niches	Promising early experimental studies	Delivery efficiency, safety, durability and manufacturing complexity

Note: IVD, intervertebral disc; NF-kappaB, nuclear factor kappa-light-chain-enhancer of activated B cells; OA, osteoarthritis; SASP, senescence-associated secretory phenotype.

8. Future directions for regenerative translation

The current evidence supports stem/progenitor cell senescence as a unifying but heterogeneous mechanism in skeletal degeneration. The strongest causal evidence exists where senescent cells have been selectively cleared or where specific stem/progenitor subsets have been functionally tested in disease models. Evidence is also strong for BMSC senescence in osteoporosis because multiple independent studies link senescence markers to osteogenic decline, adipogenic skewing, mitochondrial dysfunction, autophagy defects, inflammatory vesicle signaling, and bone-loss phenotypes. In osteoarthritis and intervertebral disc degeneration, evidence is rapidly expanding but remains more compartment-specific and model-dependent.

Several priorities should guide future research. First, senescence should be measured with multi-parameter panels rather than single markers. A useful panel would combine p16INK4a, p21, DNA-damage markers, lamin B1, SASP factors, mitochondrial function, autophagy/mitophagy readouts, and functional assays. Second, cell identity must be resolved using lineage tracing, single-cell transcriptomics, spatial profiling, and functional validation. Bulk BMSC or cartilage-cell cultures can obscure rare senescent progenitor subsets that have disproportionate niche effects. Third, disease stage and tissue compartment should be incorporated into experimental design. A senomorphic intervention that is effective in early inflammatory osteoarthritis may not reverse late-stage cartilage loss, and a cell product that survives in marrow may fail in an acidic disc.

A practical translational sequence can be defined as four checkpoints: verify the identity and senescence state of the relevant progenitor compartment; classify the niche as early inflammatory, mechanically maladaptive, metabolically hostile, or end-stage exhausted; match the intervention to the bottleneck by selecting senomorphic, senolytic, metabolic, EV, biomaterial, or cell-based strategies; and choose endpoints that test regenerative competence rather than senescence-marker reduction alone. This roadmap converts the mechanistic synthesis into a decision framework while avoiding duplication of the mechanism map in Fig. 1 and the disease-feedback scheme in Fig. 2.

Therapeutic studies need stronger translational endpoints. Demonstrating reduced senescence-associated beta-galactosidase staining in vitro is insufficient. Preclinical studies should report biodistribution, dose-response, durability, immune effects, sex differences, donor age, comorbidities, biomechanical conditions, and off-target effects. Clinical studies should integrate symptom scores with imaging, biochemical markers, cell-product quality metrics, and tissue-specific biomarkers. In osteoporosis, imaging endpoints could include bone mineral density, high-resolution peripheral quantitative computed tomography, marrow fat imaging, fracture-healing quality, and finite-element estimates of strength. In osteoarthritis, structural endpoints could include quantitative MRI cartilage thickness, T2 mapping, synovitis scores, bone marrow lesions, and biochemical markers of cartilage turnover. In disc degeneration, endpoints could include Pfirrmann grade, T2 relaxation, disc height, endplate change, inflammatory imaging, and pain-related functional scales.

Model-system harmonization is equally important. Many cell-culture studies use high oxygen tension, supraphysiological glucose, fetal bovine serum, and stiff plastic surfaces, none of which accurately reproduce aged marrow, cartilage, or disc niches. More physiologically relevant models should incorporate hypoxia, matrix stiffness, inflammatory gradients, mechanical load, and donor heterogeneity. Organoid and organ-on-chip systems may help bridge the gap between conventional cultures and animal models, particularly for joint and disc tissues where mechanical and biochemical environments are complex [[116], [117], [118], [119]]. Finally, ethical and practical issues must be considered. Older patients with comorbidities and polypharmacy may be more vulnerable to off-target toxicity, impaired wound healing, infection, or interactions with antiresorptive, anabolic, anti-inflammatory, or antidiabetic drugs. For cell and EV products, donor screening, viral safety, manufacturing reproducibility, and long-term surveillance are essential [120].

9. Conclusion

Stem/progenitor cell senescence provides a mechanistic framework for linking aging, inflammation, impaired regeneration, extracellular matrix disruption, and maladaptive mechanical signaling in osteoporosis, osteoarthritis, and intervertebral disc degeneration. Senescent BMSCs, skeletal stem cells, cartilage progenitor cells, synovial stromal cells, HSC-associated immune programs, and disc-derived progenitors influence disease through loss of regenerative competence and gain of inflammatory or matrix-catabolic functions. The field is moving from descriptive marker studies toward mechanistic, single-cell, and intervention-based evidence. Translation will depend on rigorous cell identity, standardized senescence assays, disease-stage stratification, and therapies that target the correct compartment without impairing physiological repair. A refined understanding of stem/progenitor cell senescence may ultimately support disease-modifying regenerative strategies for skeletal degeneration rather than symptomatic management alone.

CRediT authorship contribution statement

Heng Qiu: Conceptualization, Investigation, Writing - original draft. Yong Huang: Investigation, Writing - review and editing. Danni Luo: Writing - review and editing, Validation. Li Wang: Conceptualization, Methodology, Supervision, Project administration, Funding acquisition, Writing - review and editing. All authors approved the submitted version.

Ethics approval and consent to participate

Not applicable. This article is a review of published literature and did not involve human participants, human data, or animal experiments.

Data availability

No new data were generated or analyzed in this review.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

No artificial intelligence tools were used in the preparation and writing of this article.

Funding

This work was supported by the Health Care of Cadres of Sichuan Province (grant no. CGB2024049) and the Healthcare Research Program for Scientific Researchers in Sichuan Province (grant no. Chuan Gan Yan 2024-506). The article processing charge was supported by the same funding sources.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.