Bone Cell Senescence: Mechanisms and Perspectives


  • Pierre J Marie

    Corresponding author
    1. Inserm UMR-1132, Paris, France
    2. University Paris Diderot, Sorbonne Paris Cité, Paris, France
    • Address correspondence to: Pierre J Marie, PhD, Inserm UMR-1132, Hopital Lariboisière, 2 Rue Ambroise Paré, 75475 Paris Cedex 10, France. E-mail:

    Search for more papers by this author


Age-related bone loss is in large part the consequence of senescence mechanisms that impact bone cell number and function. In recent years, progress has been made in the understanding of the molecular mechanisms underlying bone cell senescence that contributes to the alteration of skeletal integrity during aging. These mechanisms can be classified as intrinsic senescence processes, alterations in endogenous anabolic factors, and changes in local support. Intrinsic senescence mechanisms cause cellular dysfunctions that are not tissue specific and include telomere shortening, accumulation of oxidative damage, impaired DNA repair, and altered epigenetic mechanisms regulating gene transcription. Aging mechanisms that are more relevant to the bone microenvironment include alterations in the expression and signaling of local growth factors and altered intercellular communications. This review provides an integrated overview of the current concepts and interacting mechanisms underlying bone cell senescence during aging and how they could be targeted to reduce the negative impact of senescence in the aging skeleton. © 2014 American Society for Bone and Mineral Research.


Cell senescence is a general process that affects most cells in the organism during aging. Several mechanisms involved in cell senescence have been identified in soft tissues. The hallmarks of aging include genomic instability, telomere attrition, epigenetic alterations, and loss of proteostasis. Cell damage, cellular dysfunction, and senescence are the responses to these primary hallmarks, and stem cell exhaustion and altered intercellular communication are the final steps that characterize the aging phenotype.[1, 2] Most of these mechanisms previously reported in many cells also occur in bone cells. Other mechanisms, such as changes in expression of local factors or cell-cell communications in niches, are more relevant to the skeleton and may contribute to bone cell aging and loss of skeletal integrity (Fig. 1). This review proposes an integrated view of the current mechanisms involved in bone cell senescence, provides some perspectives related to the hierarchical importance of these mechanisms, and suggests potential therapeutic targets in the aging skeleton.

Figure 1.

General and local senescence mechanisms contributing to bone cell senescence in the aging skeleton. Both general intrinsic senescence mechanisms and local mechanisms that are more specific to the skeletal microenvironment contribute to bone cell aging. Local mechanisms in the skeleton include, among others, alterations in the expression and release of endogenous factors (black spots), altered signaling pathways and cell-cell communications, and signal disruptions in niches, leading to altered bone matrix formation and composition.

The Aging Skeleton: A Result of Complex Mechanisms

In recent years, significant advances have been made in the understanding of the role of genetic, environmental, and endogenous mechanisms involved in the alterations of skeletal integrity and fracture risk with aging.[3] The age-related bone loss after middle age results mainly from decreased bone formation relative to bone resorption. Studies in mice and humans have shown that the relative decline in bone formation with age results from processes that are linked to reduced proliferation and differentiation of multipotent stem cells (MSCs) within the bone marrow stromal cell population (also known as bone marrow-derived mesenchymal stem cells and skeletal stem cells) into osteoblasts, decreased osteoblast function, and increased apoptosis of more mature osteoblasts,[4] resulting in decreased bone matrix production. The structural components of the bone matrix also change with age with altered stoichiometry of matrix components by aging osteoblasts and increased matrix degradation,[5] possibly related to the local release of matrix by senescent cells,[6] leading to larger apatite crystals and increased stiffness. These age-related changes are believed to arise as a consequence of both intrinsic and extrinsic mechanisms.[7-9] Extrinsic mechanisms include sex-hormone deficiency, decreased physical activity, nutritional deficiency, glucocorticoid excess, and alcohol or smoking consumption. Intrinsic mechanisms include more general mechanisms of cell senescence such as alterations in cellular damage and repair mechanisms, which impact skeletal progenitor cell renewal and differentiation and bone matrix production and composition. As outlined below, these intrinsic mechanisms are of paramount importance in the altered bone cell functions and sekletal integrity associated with aging.

General Mechanisms of Cell Senescence: Impact on Bone Cells

Telomere shortening impacts bone cell differentiation

An important mechanism that contributes to cell senescence, characterized by irreversible cycle arrest, is telomere shortening. Telomere length is maintained by telomerase and protein complexes composed of shelterin and other proteins that regulate telomere maintenance by telomerase,[10, 11] which allows cells to extend telomere length and to maintain self-renewal and stem cell population. Telomere shortening has been involved in several age-related diseases and premature aging syndromes.[12] Age-related shortening of telomeres leads to reduced cell replication, which contributes to impaired DNA repair and to accumulation of damaged cells. In osteoblast progenitor cells, this leads to limited stem cell pools and impaired differentiation into bone-forming cells. The role of telomeres in skeletal aging is supported by the finding that mice with invalidated telomerase reverse transcriptase (Terc) show decreased bone mass associated with reduced osteoblast differentiation and function[13] and enhanced osteoclastogenesis resulting from a proinflammatory osteoclast-activating microenvironment.[14] Mechanistically, the reduced osteoblast differentiation in Terc−/− mice involves increased p53 expression, whose activation induces cell cycle arrest, apoptosis, and reduced Runx2 expression.[15] Thus, telomere dysfunction limits osteoblast differentiation of skeletal progenitor cells and contributes to age-related bone loss (Fig. 2). Consistent with this concept, telomerase overexpression in human mesenchymal stem cells (MSCs) was found to maintain differentiation in vitro[16-18] and to enhance bone formation in vivo.[19, 20] This suggests that increasing telomerase expression could prevent telomere shortening and replicative senescence in osteoblast precursor cells. Recent data indicate that telomerase gene therapy in old mice delays aging without increasing cancer,[21] which may open a new area focused on telomerase reactivation to limit bone cell senescence and attenuate age-related bone loss.

Figure 2.

Integrated view of the current mechanisms involved in bone cell senescence. Current data indicate that skeletal aging results from a complex network of interacting effector programs including intrinsic cellular processes, alterations in endogenous factors and signals, and possibly inflammatory processes (dashed lines). This hierarchical basis of senescence mechanisms results in altered bone cell differentiation, function, and fate, which contribute to altering skeletal integrity.

Increased oxidative stress causes bone cell damage

Damage to cellular components induced by reactive oxygen species is an important mechanism contributing to aging. As is the case in other tissues, the accumulation of cellular and tissue damage caused by oxidative stress contributes to the adverse effects of aging on bone cells.[22] Reactive oxygen species (ROS) increase with age and cause cell damage and cell death in bone cells, supporting the concept that the increased oxidative stress plays a pathological role in the development of bone fragility with aging.[23] Mechanistically, the age-related increase in ROS and decrease in glutathione reductase activity are associated with increased phosphorylation of p53 and p66shc that regulate cell senescence and aging,[24] leading to activation of osteoblast/osteocyte apoptosis and decreased bone mass in mice.[25] This mechanism can be reproduced in mice deficient in the cytoplasmic copper/zinc superoxide dismutase (Sod1) gene, which exhibit increased intracellular ROS, skeletal fragility, and low-turnover osteopenia related to reduced osteoblastic cell proliferation and impaired osteoblast viability,[26] a mechanism that is exacerbated by mechanical unloading.[27] The age-related ROS accumulation may also contribute to the age-related shift in bone marrow stromal cell differentiation toward adipocytes instead of osteoblasts, resulting in increased bone marrow adipogenesis associated with decreased osteoblastogenesis.[28] This effect results from the activation of PPARγ signaling in osteoblastic cells by oxidized lipids.[29] Additionally, the increased endogenous glucocorticoid expression with age contributes to increased ROS production and FOXO activity,[30] which in turn antagonizes Wnt signaling.[29, 31] Both the ROS-mediated activation of FOXOs and the sequestration of β-catenin by PPARγ can account for the age-related decrease in Wnt signaling,[32] which is an important pathway that controls osteoblast and adipogenic differentiation.[33] The role of FOXOs as inhibitors of Wnt/β-catenin signaling during aging is supported by the finding that mice lacking FOXOs in skeletal progenitor cells show increased bone formation and bone mass during aging.[34] ROS are also known to stimulate osteoclastic bone resorption,[35] suggesting that the age-related increase in ROS may contribute to the promotion of bone resorption in addition to reduced osteoblastogenesis (Fig. 2). Consistently, oxidative stress was found to be associated with increased bone resorption and low bone mass in women.[36]

Although the accumulation of oxidative DNA damage contributes to age-related osteoblast dysfunction, the mechanism by which DNA damage leads to accelerated bone aging is not fully understood. Recent data have pointed out a role of the NF-κB pathway in age-related bone loss. This pathway is an important modulator of both osteoclasts[37] and osteoblasts.[38, 39] Consistent with a role of NF-κB in DNA damage, Chen and colleagues[40] showed that mice that are deficient in ERCC1-XPF, an endonuclease that is required for DNA repair, exhibit severe bone loss caused by increased NF-κB transcription in osteoblastic and osteoclastic cells. This resulted in senescence of osteoblastic cells, reduced bone formation, and increased osteoclastogenesis. These studies indicate that DNA damage induces bone cell dysfunctions in part through an NF-κB-dependent mechanism (Fig. 2).

The finding that oxidative stress in bone cells is an important mechanism underlying the age-related skeletal fragility raises the hypothesis that this mechanism may be antagonized by antioxidants. Consistently, treatment with antioxidants was found to prevent bone loss in animal models[41-43] and to abrogate the increased apoptosis in aged mice, although this effect was limited by concomitant inhibition of Wnt signaling. Similarly, treatment of SOD1-deficient mice with the antioxidant vitamin C decreased bone fragility and osteoblastic survival and attenuated bone loss during unloading.[26, 27] Intermittent parathyroid hormone (iPTH) administration was also shown to attenuate the rise in intracellular ROS, p66shc phosphorylation, and FOXO transcriptional activity induced by oxidative stress in osteoblastic cells, as well as their adverse effects on Wnt signaling and osteoblast apoptosis, indicating that this agent can antagonize the age-related increase in oxidative stress in aging mice.[44] In addtition to iPTH, estrogens and androgens exert direct antioxidant actions on bone cells, which may contribute to their bone-protective effects in mice.[45] Oxidative stress can also be targeted in osteoclasts to reduce bone loss. Osteoclasts express NADPH oxidase 4 (NOX4), an enzyme involved in ROS production, and mice deficient in NOX4 display high bone density and reduced osteoclast number. Inhibition of NOX4 leads to reduced bone resorption[46] and bone loss induced by ovariectomy in mice.[47] Thus, combating the age-associated oxidative stress in bone cells can lead to an attenuation of the adverse effects of aging on bone (Fig. 3).

Figure 3.

Potential targets to overcome bone forming cell senescence. Bone cell senescence may possibly be attenuated by agents (in italics) that target some of the major mechanisms that are involved in cell senescence. Such therapeutic approaches aimed at reducing bone cell senescence may have a positive impact on bone formation and skeletal integrity.

Epigenetics: implications on bone cell aging

Changes in epigenetic pathways such as DNA methylation, histone methylation, and histone acetylation are clearly linked to cellular senescence and aging through their control of chromatin structure and telomere length.[48, 49] In bone, several epigenetic mechanisms were found to influence the transcription of genes that control osteoclastogenesis and osteoblastogenesis. Notably, DNA methylation controls RANKL and osteoprotegerin (OPG),[50] osteocalcin,[51] BMP2,[52] sclerostin,[53] and C/EBPα.[54, 55] In addition, histone demethylases regulate the expression of H3 histones that control RANKL-induced NFATc1 expression and osteoclast differentiation.[56] Histone demethylases also regulate osteogenic differentiation[57, 58] and modulate the expression of Runx2 and Osterix,[59, 60] whereas histone methylation controls PPARγ target gene promoters.[61] Histone acetylases and deactelyases (HDACs) are highly involved in the regulation of osteoblast differentiation[62] by controlling histone H4,[63] C/EBPα,[55] and Runx2.[62, 64] Given these important effects, it is expected that disturbances in epigenetic regulators in bone cells may contribute to the pathophysiology of age-related bone loss (Fig. 2). In support of this concept, mice deficient in DNA methyltransferase 1 (Dnmt1), the most prevalent DNA methyltransferase, exhibit decreased bone mineral density without altered survival or mortality.[65] One particular group of HDACs called sirtuins play an essential role in the regulation of chromatin structure, telomere stability, senescence, DNA repair, cell differentiation, and apoptosis.[66] A member of the sirtuin family, the NAD-dependent histone deacetylase silent information regulator T1 (SIRT-1), regulates transcription factors involved in aging.[67] In bone, SIRT-1 was shown to promote osteoblast differentiation and to attenuate adipogenesis through the regulation of SOST,[68] PPARγ,[69] Runx2,[70] and β-catenin.[71] SIRT-1 also negatively controls NF-κB signaling. In mice, repression of SIRT-1 in osteoblasts or osteoclasts activates NF-κB signaling, resulting in reduced bone formation and enhanced bone resorption, a phenotype rescued by inhibition of NF-κB.[72] Interestingly, SIRT-1 overexpression protects against age-related bone loss, whereas reducing its expression causes decreased bone formation in mice,[68, 73] further indicating that SIRT-1 is an important epigenetic regulator in aging bone cells.

MicroRNA (miRNA) are other epigenetic regulators that control gene expression directly or through crosstalks with DNA methylation[49] and may thereby play an important role during aging.[74, 75] In osteoclasts, several miRNAs were shown to regulate key signaling molecules such as Notch, RANKL, RANK, calcitonin receptor, and RhoA.[76] In bone-forming cells, some miRNAs were found to enhance osteoblast differentiation in MSCs by enhancing promoting signals (such as Runx2, Osx, ATF4, Smad5, ERK, Cx43, and LRP6) or by suppressing inhibitory signals (such as Foxo1, DKK1, and PPARγ).[77, 78] Moreover, miRNA were found to regulate HDACs that control osteoblast differentiation and bone formation.[79, 80] In support of a role of miRNA in bone cell functions in the aging skeleton, dysfunctions in miRNA-dependent mechanisms have been reported in osteoporosis.[76]

Given the importance of epigenetic mechanisms in the regulation of bone cells, it may be of interest to target these processes to optimize bone cell differentiation or function during aging. For example, recent data suggest that miRNAs can be targeted to promote osteoblast differentiation and bone formation and repair in vivo.[78, 81, 82] In contrast, it is still uncertain whether targeting HDACs may have beneficial effects on bone cells. Indeed, HDAC inhibition was found to inhibit or increase osteoclast differentiation in vitro or bone resorption in vivo, presumably because of the distinct role of specific HDACs in bone.[62] Similarly, HDACs inhibitors were shown to stimulate osteoblast differentiation in vitro but this has not been confirmed in animals, possibly because these inhibitors promote apoptosis in immature osteoblast progenitor cells.[83] Pharmacological enhancement of SIRT-1 could be one interesting option to promote osteogenic differentiation. Indeed, increasing the activity of the SIRT-1 protein by the phytoestrogen resveratrol in MSCs resulted in increased osteoblast differentiation and decreased adipocyte differentiation in vitro.[70, 84, 85] Resveratrol also inhibits NF-kB activity and osteoclast differentiation in vitro.[84, 86] In vivo, resveratrol administration was found to preserve bone mineral density in elderly mice, in part through the attenuation of the age-associated increase in oxidative stress.[87] In a mouse model of progeria, resveratrol treatment improved trabecular bone structure and mineral density through enhanced binding between SIRT-1 and lamin A.[88] Overall, these studies suggest that SIRT-1 could be targeted to alleviate some of the senescence mechanisms in bone cells that contribute to skeletal aging. Further studies in humans are needed to identify specific epigenetic changes associated with aging that could be targeted to counteract the age-related dysfunctions in bone cells.

Decreased autophagy and proteasome degradation: impact on bone cell differentiation and function

The maintenance of cellular homeostasis requires efficient proteolysis of damaged proteins. Autophagy is a lysosome-dependent degradation pathway whose primary role is to break down damaged cellular components. Under stressful conditions, autophagy is increased, which helps cells to survive stress through increased recycling of cellular components. Thereby, autophagy plays a critical role in the maintenance, differentiation, and survival of stem cells as well as more differentiated cells.[89, 90] Autophagy was recently identified as a new effector of senescence.[91] Autophagic efficiency declines with age,[92] which may contribute to the age-related accumulation of damaged molecules and organelles induced by starvation and ROS in aging cells.[93, 94] In the skeleton, all bone cells exhibit autophagy[95] and current evidence indicates that this process plays a role in bone physiology and pathology[96] (Fig. 2). In bone-resorbing cells, autophagy in response to oxidative stress induces osteoclast precursor cell differentiation.[97] In addition, autophagy proteins participate in the secretion of lysosomal contents into the extracellular space, an important mechanism involved in bone resorption.[98] However, the actual importance of autophagy in regulating osteoclastogenesis and function during aging remains unclear.[96] In contrast, a role of autophagy in osteoblastogenesis is supported by the finding that specific deletion of FIP200, an essential component of mammalian autophagy, leads to defective osteoblast differentiation, reduced bone formation, and osteopenia in mice.[99] Although the role of autophagy in osteocytes during aging is not fully established, it was recently found that conditional deletion in osteocytes of Atg7, a gene essential for autophagy, leads to decreased bone formation and bone mass associated with increased oxidative stress,[100] suggesting that suppression of autophagy in osteocytes induces a phenotype resembling aging. Thus, current data suggest that a decline in autophagy with age may contribute to bone aging. Whether autophagy may be targeted to reduce bone cell senescence and bone loss associated with aging remains to be determined.

Another important mechanism regulating intracellular protein degradation is the ubiquitin-dependent proteolysis system (UPS).[101] This system mediates the ubiquitination of proteins through the actions of ubiquitin ligases, which direct their degradation through the proteasome. The UPS-mediated clearance of damaged and short-lived proteins ensures a protein quality control, a mechanism that impacts cell proliferation, differentiation, and apoptosis.[102] Because of its important function, a decrease in proteasome activity with age may lead to the accumulation of oxidized and damaged proteins,[103-105] which are directly linked to aging.[106] In bone cells, significant advances have been made in our understanding of the role of UPS and E3 ubiquitin ligases in the proteasomal degradation of key regulatory proteins such as receptor tyrosine kinases (RTK), signaling molecules, and transcription factors that control osteoclast and osteoblast differentiation and function.[107] Because the ubiquitination/degradation system controls many processes in bone cells, this mechanism could be targeted to improve bone cell function (Fig. 3). For example, proteasome inhibitors were shown to increase MSC osteogenic differentiation and bone formation in mice.[108, 109] A more specific approach for promoting bone formation may be to target E3 ubiquitin ligases that control proteasome degradation.[107] In support of this concept, it was shown that inhibiting the activity of E3 ligase c-Cbl in murine and human MSCs promoted osteoblast differentiation and survival through activation of RTK signaling[110] and STAT5-Runx2 transcriptional activity.[111] Whether targeting E3 ubiquitin ligases or other proteins controlling proteasome degradation in bone cells may attenuate age-related bone cell senescence warrants further investigation.

Senescence Mechanisms in the Skeletal Microenvironment

Alterations of signaling molecules: impact on bone cell fate

Some aging mechanisms are highly relevant to the bone microenvironment because they involve changes in the expression and signaling of local factors expressed by bone cells and/or present in the bone matrix (Fig. 2). In mice, changes in transforming growth factor-β (TGF-β) and bone morphogenetic protein (BMP) signaling correlate with the reduced osteoblastogenesis and increased adipogenesis with age.[112] Insulin-like growth factor-1 (IGF-1) levels in the bone matrix were also shown to decline with age in mice, which can account for the reduced bone formation during aging.[113] Because Wnt signaling is an important pathway that regulates bone homeostasis,[114] changes in Wnt signaling may impact skeletal integrity. In mice, the expression of Wnt3a, Wnt5a, and Wnt10b, which are important ligands controlling osteoblast and adipocyte differentiation,[61, 115] markedly decrease with age.[116] Consistently, loss of Wnt10b in mice results in an age-progressive osteopenia attributable to the loss of multipotent progenitors.[117] Thus, the decrease in Wnt signals with age likely contributes to the altered bone formation in aging mice. Recently, an unexpected shift from canonical to noncanonical Wnt signaling owing to elevated expression of Wnt5a in aged hematopoietic stem cells was found to cause stem cell aging in mice.[118] Whether such a mechanism may also affect osteoblastic stem cells during aging needs to be investigated.

With age, there is an inverse relationship between bone marrow adiposity, which increases, and osteoblastogenesis, which decreases, that impacts bone density and strength.[119] This effect seems to result from the switch in the differentiation of a common progenitor cell to adipocytes rather than to osteoblasts. The balance between osteoblast and adipocyte differentiation is regulated in a complex manner by multiple hormonal and growth factors, signaling mechanisms, and transcription factors,[120, 121] which may be complicated by the impact of local adipokines secreted locally by adipocytes.[122] An essential transcription factor that triggers adipocyte differentiation is PPARγ2.[123] PPARγ2 expression was shown to increase with age in mice.[112] This may explain in large part the excessive bone marrow adipocity and reduced osteoblastogenesis associated with aging, providing a potential target to counteract this age-related phenotype. Indeed, genetic or pharmacological suppression of PPARγ expression or activity results in increased osteoblast differentiation at the expense of adipogenesis.[124-127] Negative regulators of PPARγ2 expression such as growth hormone,[128] iPTH,[129] strontium,[130] fibroblast growth factor-2,[131] TGF-β,[132] Wnt signaling,[133, 134] vascular endothelial growth factor (VEGF),[135] and mechanical loading[136, 137] and its downstream signal mTORC2[138] were all shown to promote osteoblastogenesis at the expense of adipogenesis in cultured MSC and in vivo. Whether inhibition of PPARγ activity may reallocate bone marrow MSCs to osteoblast rather than to adipocyte differentiation during aging in humans deserves further investigation.

Recent studies have revealed that osteocytes control bone cells through the release of factors such as sclerostin, the product of the SOST gene.[139] Osteocytes might also play a role in bone resorption via RANKL expression.[140-142] Whether the expression of these osteocyte-derived molecules changes with age and may affect osteoblast/osteoclast differentiation and/or function during aging is not yet known. Importantly, however, osteocytes can sense mechanical loading and translate strain into physiological signals through multiple mechanisms including integrin-matrix mediated signaling, channels activity, and Wnt and mTORC2 signaling.[143] With age, these signals are expected to decrease when physical activity/mechanical strain declines, and this may contribute to the alteration of the balance between bone resorption and formation. Interestingly, recent data have linked the extracellular matrix (ECM) and lamin A, a protein of the nuclear membrane, to mesenchymal cell fate. Increasing ECM matrix stiffness leads to increased lamin A expression, enhanced osteoblast differentiation, and reduced adipogenic differentiation in MSCs.[144] Mechanistically, lamin A was shown to downregulate adipocyte differentiation and to promote osteoblastogenesis in MSCs in part by regulating RUNX2 and PPARγ[144-148] and endogenous VEGF.[135] It is thus conceivable that disturbances in the mechanisms linking ECM stiffness, nuclear proteins, and mesenchymal cell fate could contribute to the altered osteoblast/adipocyte differentiation during aging.

Altered intercellular communications during aging: a question of niche?

In many tissues, stem cell proliferation, self-renewal, and differentiation are highly dependent on cell-cell interactions and signaling mechanisms.[149] <zaq;1> In the bone marrow, cellular interactions are influenced by aging.[150, 151] Osteoblasts are interconnected by gap junctions and hemichannels, such as connexin 43 (Cx43), that play a critical role in normal osteoblast differentiation, function, and apoptosis.[152-154] In vivo, reduction in Cx43 function results in osteopenia in mice,[155] although a dominant negative Cx43 mutant protein may also increase osteoblast function,[156] indicating that the signals generated by these cell-cell communications are essential for the maintenance of skeletal integrity. Interestingly, aging is associated with an impairment in the capacity of osteoblastic cells to generate functional gap junctions in response to iPTH.[157] In osteocytes, Cx43 is involved in the skeletal response to mechanical forces[158] and is required for osteocyte survival.[159] Recent data indicate that oxidative stress reduces Cx43 expression and thereby contributes to osteocyte apoptosis, suggesting that Cx43 channels may protect osteocytes against oxidative stress-induced cell death during aging.[160] Moreover, integrin α5β1, an important integrin regulating bone-forming cells and mechanotransduction,[161] can interact directly with Cx43, and this interaction is required for mechanical stimulation-induced opening of the Cx43 in osteocytes.[162] Thus, alterations in the expression or function of these connecting molecules may contribute to bone loss associated with aging (Fig. 2).

Osteoblasts are also connected via the cell-cell adhesion molecules cadherins, which control osteoblast differentiation, activity, and survival.[163, 164] Cadherins interact with the canonical Wnt/β-catenin signaling,[165] and it was shown that the interaction between N-cadherin (Cadh2) and the Wnt co-receptors LRP5/6 controls osteoblast differentiation and function.[166] Disruption of Cadh2 function favors adipogenic differentiation of bone marrow stromal cells,[167] and we recently reported that Cadh2 controls the switch from osteoblast to adipocyte lineage differentiation during aging through changes in endogenous Wnt5a and Wnt10b in osteoblasts. Interestingly, it was found that blocking the CADH2-Wnt interaction can increase Wnt5a and Wnt10b expression, bone formation, and bone mass in senescent osteopenic mice, suggesting that CADH2 may be a potential target for promoting bone formation and bone mass in the aging skeleton (E Haÿ and colleagues, unpublished data, 2014).

Stem cell niches are defined as physiological entities consisting of supporting cells, diffusible and cell surface signaling molecules, as well as physical parameters such as matrix rigidity.[168, 169] Disruption in the signals within the niche may contribute to aging,[170] but the underlying mechanisms that modulate niche cells during aging are not yet known. In bone, the hematopoietic stem cell (HSC) niche is a complex entity composed of multiple cell types, including osteoprogenitor cells.[171-173] Osteoprogenitor cells (skeletal stem cells such as pericytes) appear to be the main actors that support HSC replication and differentiation in the hematopoietic niche.[174, 175] This effect is mediated by cell-cell interactions involving molecules such as Cadh2[176-178] and Cx43.[179] By analogy, it can be hypothesized that age-related alterations in the interactions between mature osteoblasts and MSCs, or in the signals produced by these cells in their niche, may alter MSC renewal, differentiation, or function, thus contributing to the age-related decrease in bone formation (Fig. 1). Given that the niche is a dynamic microenvironment that changes during physiologic states,[180] changes in the niche components or in transmitted signals between cells during aging may determine the replacement hierarchy of MSCs within the niche. Therefore, identifying the changes in niche components and signals during aging may provide novel clues about the possible approaches that may be used to restore MSC expansion and differentiation in the aging skeleton.

Conclusions and perspectives

Our current knowledge indicates that cellular senescence is a collective phenotype composed of complex networks of effector programs. Bone cell aging arises from several interacting processes involving intrinsic senescence mechanisms, alterations in local factors and signaling pathways, and signal disruptions in osteogenic niches, providing a hierarchical basis of senescence mechanisms (Figs. 1 and 2). Based on this knowledge, it is now apparent that some of these mechanisms may be targeted with a significant impact on bone formation and skeletal integrity (Fig. 3).

Although important progress has been made in our understanding of the mechanisms underlying bone cell senescence, several important issues remain to be addressed. First, as in other tissues, the absence of good candidate markers of senescent cells in vivo is a major drawback.[181] Also, in vitro senescence may not model in vivo aging. Therefore, the availability of reliable senescent markers in vivo are needed to predict the onset and evolution of senescent cells in the aging bone, which could help prevent the damaging processes in bone cells. Second, recent studies indicate that cellular senescence leads to the secretion of proinflammatory proteins that contribute to the chronic inflamation associated with aging.[182] In bone, senescent cells and preadipocytes may acquire a pro-inflammatory phenotype and produce cytokines/chemokines with a negative impact on bone homeostasis (Fig. 3). Thus, the role of this inflammatory state in the aging skeleton needs to be investigated. Finally, a key issue will be to determine, among the numerous senescence processes identified in bone cells, what would be the best target for optimally preserving the number and function of bone-forming cells. In the future, the answers to these questions may provide new insights into potential therapies for attenuating bone cell senescence and age-related bone loss.


The author states that he has no conflicts of interest.


The work of the author was supported by grants from the Ministère de la Recherche, the Agence Nationale de la Recherche, the FP7 European Commission Program, the Société Française de Rhumatologie, and the Association Rhumatisme et Travail (Paris, France). The author thanks Dr Zuzana Saïdak for language correction of the manuscript.