Rejuvenation of Senescent Mesenchymal Stem Cells to Prevent Age-Related Changes in Synovial Joints

Abstract

Mesenchymal/medicinal stem/signaling cells (MSCs), well known for regenerative potential, have been involved in hundreds of clinical trials. Even if equipped with reparative properties, aging significantly decreases their biological activity, representing a major challenge for MSC-based therapies. Age-related joint diseases, such as osteoarthritis, are associated with the accumulation of senescent cells, including synovial MSCs. An impaired ability of MSCs to self-renew and differentiate is one of the main contributors to the human aging process. Moreover, senescent MSCs (sMSCs) are characterized by the senescence-messaging secretome (SMS), which is typically manifested by the release of molecules with an adverse effect. Many factors, from genetic and metabolic pathways to environmental stressors, participate in the regulation of the senescent phenotype of MSCs. To better understand cellular senescence in MSCs, this review discusses the characteristics of sMSCs, their role in cartilage and synovial joint aging, and current rejuvenation approaches to delay/reverse age-related pathological changes, providing evidence from in vivo experiments as well.

Introduction

Mesenchymal/medicinal stem/signaling cells (MSCs) are multipotent progenitor stem cells derived from adult tissues/fluids such as bone marrow ¹ , adipose tissue ² , skeletal muscle tissues ³ , synovial membrane ⁴ , and fluid ⁵ , urine ⁶ , and others. In fact, MSCs belong to highly heterogeneous cell populations. Therefore, their biological features can vary between individuals, tissues, or even within a single cell colony ⁷ . Introducing such ambiguously characterized cells into clinical trials can be complicated and may jeopardize the safety of MSC-based therapies. Consequently, a single set of rules allowing for the identification of human MSCs was created and approved in 2006 by the International Society for Cellular Therapy. The minimum criteria to define unique populations of human MSCs include the following: (1) plastic-adherence properties; (2) the expression of CD73, CD90, and CD105, and absence of CD14, CD19, CD34, CD45, and HLA-DR–specific surface markers; and (3) capability of multilineage differentiation in vitro ⁸ . Furthermore, MSCs are immune evasive due to the low expression of major histocompatibility complex (MHC) class I and the absence of MHC class II molecules (at the early passage) ⁹ . Their regenerative potential is principally based on the secretion of bioactive molecules. A cocktail of various growth factors, cytokines, chemokines, and enzymes has a reparative, anti-inflammatory, and immunomodulatory effect^10,11. Noteworthily, the composition of secretome strictly depends on the MSCs’ origin and thus varies markedly between MSCs from different sources ¹² .

The interest of scientists worldwide is enormous due to the remarkable MSCs’ therapeutic potential executed during the regeneration of various damaged tissues. Some of these defects are the result of injury, while others are associated with aging. Generally, aging is a continual and irreversible process, which affects all cells, including MSCs. Different internal^13,14 and external factors ¹⁵ are responsible for the induction and/or acceleration of cell aging. Aging cells progressively lose their reparative and regenerative characteristics leading to increased tissue and cellular dysfunction. As an example, stress-induced accumulation of senescent cells, especially senescent chondrocytes and synovial MSCs (S-MSCs), can ultimately lead to the development of joint-related diseases such as osteoarthritis (OA). Recently, several studies showed that the removal of these cells from the joint enhanced the self-renewal and regeneration of cartilage ¹⁶ . Overall, the transplantation of young MSCs can be an innovative tool to prevent or slow the onset of age-related conditions. On the other hand, MSCs’ therapeutic efficacy may be skewed in an aging environment ¹⁷ . Therefore, optimizing strategies combating aging and cellular senescence is vitally important to minimize the pathological effects of age-related processes. In this review, we describe the characteristics of senescent MSCs (sMSCs), the factors causing senescence, and the role of senescent chondrocytes and sMSCs in cartilage and the synovial joint. We also provide a summary of current rejuvenation techniques.

Senescence at a Cellular and Molecular Level

MSCs, similarly to other cell types, go through age-related changes and become senescent over time. Indeed, both intrinsic [eg, reactive oxygen species (ROS)] and extrinsic stressors (eg, environmental toxins) are principal triggers of the aging process^18,19. SMSCs are characterized by an irreversible cell growth arrest despite being metabolically active. Other identifiers of senescence are changes in cell shape, decreased proliferation, impaired differentiation ability, and reduction in capacity to form colonies^14,20. In addition, sMSCs show elevated lysosomal β-galactosidase (SA-β-gal) levels compared with cells in early passages due to the increased lysosomal mass and altered pH^21,22. SA-β-gal activity is commonly used as a senescence marker, but in some cases, it may produce false data. Of note, it cannot be detected in senescent cells with a defective lysosomal β-galactosidase ²³ . Therefore, appropriate identification of senescent cells requires a combination of senescence markers. Alpha-fucosidase (SA-α-Fuc), so-called glycosidase, is another lysosomal enzyme observed in many senescent cells regardless of the senescence-inducing stimulus ¹⁴ . It is considered to be a more accurate senescence marker compared with SA-β-gal ²⁴ . However, there are still doubts about its specificity when it comes to the identification of sMSCs.

The expression of surface molecules undergoes dynamic changes during the aging of MSCs. Although these changes depend on the conditions under which MSC aging occurs. The reduced expression of MSCs’ markers, such as CD146 (known as melanoma cell adhesion molecule, MCAM) ²⁵ , stromal cell surface marker-1 (STRO-1) ²⁶ , or CD106 (known as VCAM-1) ²⁷ , was detected during in vitro cultivation and in aged organisms. Moreover, the enhanced expression of CD264 was associated with sMSCs only during the long-term culture in vitro. Notably, CD264-positive MSCs also show increased SA-β-gal activity and reduced proliferation and differentiation ability. This suggests that CD264 may be considered another indicator of senescence ²⁸ . SMSCs are also characterized by a senescence-messaging secretome (SMS), typically manifested by the release of molecules [interleukin (IL)-1α, monocyte chemoattractant protein (MCP)-1, insulin-like growth factor binding protein 4 (IGFBP4), etc] with an adverse effect on adjacent cells and their microenvironment ²⁹ . Typical features of sMSCs are depicted in Fig. 1.

Figure 1.

Hallmarks of senescent mesenchymal stem cell (sMSC). Cell cycle arrest occurs due to the action of internal and environmental stressors that impair the proliferative capacity. Consequently, MSC acquires a senescent phenotype, which is manifested by an enlarged and flattened morphology, the presence of surface molecules associated with senescence, enlarged lysosomes with a raised level of enzymes (α-Fuc and β-gal), increased vacuolar volume, elevated content of mitochondrial DNA (mtDNA), and accumulation of DNA damage causing changes in gene expression. Subsequently, MSC secretory activity is shifted toward the senescence-messaging secretome (SMS). This is related to the increased release of factors (cytokines, growth factors, etc) and particles (eg, exosomes). The secreted molecules and particles can operate as paracrine factors triggering senescence in neighboring nonsenescent cells or autocrine factors augmenting the senescent phenotype within the same cell. Furthermore, the presence of dysfunctional mitochondria is linked to the massive production of ROS that further amplify the senescent status. IFN: interferon; IGFBP: insulin-like growth factor binding protein; IL: interleukin; MCP: monocyte chemoattractant protein; MMP: matrix metalloproteinase; mtDNA: mitochondrial DNA; ROS: reactive oxygen species; SA-α-Fuc: senescence-associated lysosomal α-l-fucosidase; SA-β-gal: senescence-associated beta-galactosidase; TIMP: tissue inhibitor of metalloproteinases.

To assess aging in MSCs, numerous assays evaluating cell morphology and density, proliferative potential, mitochondrial function, and secretome composition can be applied ³⁰ . Experimentally, senescence is determined by measuring the activity of enzymes mentioned above ²¹ and length of telomeres ³¹ , monitoring the expression of selected markers (p16^INK4a, p21^Cdkn1a, and p53) ³² and alterations in gene methylation and epigenetic processes^33,34, etc. In the following section, the cellular and molecular attributes of sMSCs will be discussed in more detail.

Morphology

SMSCs are characterized by a flattened, enlarged, and irregular shape. They also display granular cytoplasm with many cell inclusions, vacuoles, and lysosomes ¹⁴ . These exceptional morphological features are often used for sMSC recognition. Accordingly, it is possible to predict the biological behavior of MSCs under specific conditions. For example, Block et al. demonstrated that bone marrow–derived mesenchymal stem cells (BM-MSCs) isolated from elderly donors consisted of a small and a large cell subpopulation. Interestingly, the subpopulation of small-sized BM-MSCs possessed a similar level of ATP as BM-MSCs derived from young donors. The subpopulation of large BM-MSCs exhibited a decreased amount of ATP compared with BM-MSCs derived from young donors and a small subpopulation of BM-MSCs isolated from elderly people ³⁵ . Besides altered cell size, senescence is accompanied by changes in the number, shape, and size of some organelles (eg, mitochondria, vacuoles, and nucleus). A typical characteristic of sMSCs is senescence-associated vacuole formation ²¹ . However, it still needs to be discovered what their exact cellular function is. The aging process is generally associated with modifications in cytoskeletal protein structure and/or expression. This causes disruption of the cytoskeleton architecture, which, in turn, leads to the morphological transformation of the cell membrane or organelles. Notably, the actin structure of senescent BM-MSCs was confirmed to be poorly organized, decreasing cell motility drastically. Behind the less dynamic cytoskeletal structure and slow actin turnover can, at least partially, stand the upregulated expression of actin-binding proteins, such as cortactin, α-actinin 2, and adducins, supporting the actin cytoskeleton stability and increasing cell rigidity. At the same time, a significant downregulation of proteins controlling microtubule dynamics [microtubule-associated proteins MAPRE1 (EB1), MAP1B, and MAP4] was detected in senescent BM-MSCs, indicating the extensive cell cytoskeleton rearrangement with senescence ³⁶ . Moreover, the membranes of senescent cells exhibit altered expression of a wide range of proteins. For instance, caveolin-1 upregulation is linked to morphological alterations in senescent BM-MSCs as well^37,38. Impaired mitochondrial dynamics (mtDNA damage, increased ROS production, reshaping, fusion, or fission) can also contribute to the altered mitochondrial morphology in sMSCs. Intriguingly, the process of mitochondrial fusion increases, whereas the fission decreases with senescence in BM-MSCs ³⁹ . Conversely, Barilani et al. ⁴⁰ detected a reduction in mitochondrial mass. The content of mtDNA, however, was elevated in aging BM-MSCs.

DNA Damage

DNA damage, if not removed by DNA reparative mechanisms, is one of the main factors that accelerate cell aging. DNA is frequently damaged as a consequence of increased oxidative stress. ROS are generated due to cellular metabolic pathways, and their low concentrations are necessary for the normal proliferation and differentiation of MSCs^41,42. Although high ROS production is associated with sMSCs, confirmed in MSCs derived from bone marrow ⁴³ , umbilical cord blood (UC-MSCs) ⁴⁴ , and Wharton’s Jelly (WJ-MSCs) ⁴⁵ . Incubation with N-acetylcysteine can prevent MSCs from entering the senescent stage as it serves as an oxygen scavenger and decreases the subsequent DNA damage ⁴⁶ . Another option is the excessive stimulation of specific DNA repair pathways since their downregulation was identified in senescent human BM-MSCs ⁴⁷ . Some studies have shown that spontaneous malignant transformations may occur during the long-term culture of MSCs ⁴⁸ . Even though the study by Wang et al. ⁴⁹ evidenced that long-term cultured human UC-MSCs developed genomic alterations, the malignant transformation was not confirmed.

Chromatin Reorganization and Epigenetic Modifications

DNA methylation and histone modifications (eg, acetylation) are the primary control mechanisms involved in chromatin remodeling. Epigenetic alterations are often found in aging MSCs and sMSCs. Changes in methylation were mainly observed in CpG sites in the promoter regions of homeobox genes and genes implicated in differentiation. The reduced expression of the homeobox gene distal-less homeobox 5, due to hypermethylation, supposedly led to impaired differentiation of BM-MSCs during long-term cultivation ³³ . Furthermore, DNA methyltransferases (DNMTs) were shown to have an essential role in regulating senescence in UC-MSCs. These enzymes control not only the methylation status of DNA but also histone marks at genomic regions of polycomb group genes (PcGs)-targeting miRNAs and p16^INK4a and p21^CIP1/WAF1 promoter regions. DNMT inhibition triggered senescence via upregulation of p16^INK4a and p21^CIP1/WAF1 expression ⁵⁰ . In addition, histone modifications are commonly present in sMSCs. As an example, histone deacetylase (HDAC) activity seems crucial for the self-renewal of adipose tissue–derived mesenchymal stem cells (AD-MSCs) and UC-MSCs. Properly functioning HDAC is presumably implicated in the balanced expression of PcGs and the jumonji domain-containing protein 3. Thus, senescence and loss of MSC self-renewal properties are related to the downregulation of HDACs, which ultimately leads to increased p16^INK4a expression and elevated SA-β-gal activity ⁵¹ . Of important note, a different form of chromatin structure called senescence-associated heterochromatin foci was observed in sMSCs. The study by Gu et al. demonstrated that the nuclei of UC-MSCs in late passages (p11 and p17) displayed chromatin localized in small and condensed spots. These represent transcriptionally inactive chromatin, further underlining the senescence-associated attenuation of gene expression ⁵² .

Telomeres

Telomere shortening, another hallmark of sMSCs, occurs continuously from the first cell passage. Telomeres shorten with each round of cell division, eventually being too short, which induces cell cycle arrest. A recent study has found that the rejuvenation of sMSCs by overexpression of telomerase reverse transcriptase (TERT) results in telomere elongation in AD-MSCs and WJ-MSCs. TERT upregulation did not affect the karyotype ⁵³ . Importantly, telomerase activity is related to sirtuins (SIRT), a group of deacetylases. Loss of SIRT1 in young MSCs stimulated cellular senescence and impaired cell proliferation, while its overexpression in aged MSCs reversed the senescent phenotype and activated cell proliferation. SIRT1-associated antiaging effects were mediated through the protection from age-related DNA damage, induction of TERT expression, and stimulation of telomerase activity. Moreover, SIRT1 upregulated the expression of tripeptidyl peptidase 1, which helps protect chromosome ends from DNA damage. Hence, the increased amount of SIRT1 can expand the cell lifespan and reverse the senescence in rat BM-MSCs ⁵⁴ . Techniques detecting DNA ends in sMSCs could allow monitoring of their aging. On the other hand, it is crucial to take into account that telomere length can be variable and vary between individuals. Therefore, relying only on the measurement of this marker may not be sufficient.

Senescence-Messaging Secretome

From the initial signs of aging in cell culture to reaching complete senescence, it lasts approximately 10 days to 6 weeks. This process depends on the presence of inducers driving the cells to senescence as well as the type of cells. Almost 30%–70% of senescent cells are reported to develop a heterogeneous SMS, which is characterized by release of the immunomodulatory, pro-inflammatory, and pro-apoptotic cytokines [IL-1, IL-6, IL-8, IL-13, IL-15, interferon (IFN)-β, transforming growth factor β (TGF-β1), etc], chemokines (eotaxin 1-3, MCP-1/CCL2, MCP-2/CCL8, RANTES/CCL5, etc), tissue-destroying proteases such as matrix metalloproteinases (MMP1, MMP3, MMP9), ligands (eg, Fas ligand), growth factors [insulin-like growth factor 1 (IGF-1), IGFBP4, IGFBP7, hepatocyte growth factor (HGF), etc], and others^55–58. Cells with SMS release even other factors involved in tissue necrosis, systemic inflammation, stem and progenitor cell dysfunction, fibrosis, and the spread of senescence to nonsenescent cells^59,60. The composition of SMS can vary greatly and mainly depends on the cell type, inductors that trigger senescence (mechanical stress, chemotherapy, radiotherapy, etc), and the metabolism of the senescent cells. For example, senescence-induced mitochondrial dysfunction is characterized by SMS but without secretion of pro-inflammatory factors such as IL-1^61,62. In fact, SMS is involved in the propagation of senescence in a paracrine and autocrine fashion ⁶³ . However, each type of senescence is stimulated by different SMS factors. TGF-β family members are mainly involved in the senescence of neighboring cells. Their downstream targets affect ROS production and the DNA damage response ⁶⁴ . The findings of Acosta et al. ⁶⁵ confirmed that some secreted factors, such as IL-1, can be involved in both autocrine and paracrine senescence. Accumulating only a small number of senescent cells can be harmful and may promote various age-related diseases^66,67. Intraperitoneal transplantation of senescent AD-MSCs into young mice induced persistent physical dysfunction and spread of cellular senescence to host tissues ⁶⁸ . Furthermore, the level of IGFBP4 has been shown to increase significantly in rat BM-MSCs with age. High IGFBP4 expression was associated with decreased osteogenic potential. This was also confirmed by IGFBP4 knockdown, which restored the osteogenic potency of aged BM-MSCs ⁶⁹ . In addition, it has been reported that IGFBP4 and IGFBP7, harvested from the conditioned media of senescent BM-MSCs, are inevitable in the induction of senescence in young BM-MSCs ⁵⁷ .

Aging Joint Environment and the Role of sMSCs

Stiffness, pain, and impaired joint mobility in older age are caused by a decrease in the amount of synovial fluid (SF), which ensures lubrication and smooth joint movement ⁷⁰ . Due to its lack, mechanical friction in the end parts of the bones occurs, resulting in thinning of the cartilage surface layer. Moreover, the loss of chondrocytes and remodeling of extracellular matrix (ECM) proteins also accompany this process. On the other hand, water content increases with age-related diseases, especially in the early stages of OA^71,72. A characteristic feature of aging cartilage is the development of excessive collagen cross-linking, leading to cartilage stiffness and impaired function ⁷³ . Cartilage disruptors include enzymes that cause the loss of collagens (mainly type II) and aggrecans^74–76. Degradation of collagens occurs due to increased levels of MMPs, particularly MMP13 ⁷⁷ . Eventually, all these factors, including metabolic and oxidative stress, disturb the cartilage environment. Thus, the aging of joints is reflected by the changes within SF composition. As an example, hyaluronic acid (HA) is the fundamental microenvironmental component of articular cartilage^78,79. It has many physiological functions (lubrication, shock absorption, stabilization of joint structure, and regulation of water balance) and is involved in various cellular processes (eg, differentiation and proliferation)^79,80. In the joint cavity, HA is synthesized by type B synoviocytes via the enzyme hyaluronan synthase (HAS), which exists in three isoforms. These produce HA polymers of various sizes. Therefore, healthy SF contains a mixture of different-sized HAs, low and high molecular weight (MW) ones ⁷⁹ . With aging, the HA concentration and MW decrease. Reduction of high MW HA seems to be vitally important since these large molecules are primarily involved in boundary lubrication at articular cartilage–cartilage interfaces^70,81. HA can interact with three main classes of cell surface receptors: (1) CD44 (a membrane glycoprotein), (2) receptor for hyaluronate-mediated motility, and (3) intercellular adhesion molecule 1. However, CD44 is the most widely distributed cell surface receptor recognized for HA binding. Some studies suggest that imbalanced HA polymers, with dominant low MW forms, may have a distinct biological impact on cellular and molecular pathways ⁸² . When high MW HA interacts with CD44, the pathway promoting the production of anti-inflammatory cytokines is activated. But when the low MW HA binds, the CD44-mediated signal has the opposite effect ⁸³ . Also, Asari et al. ⁸⁴ evidenced that low MW HA was involved in the induction and amplification of inflammation.

Undoubtedly, cartilage aging is closely related to structural changes in tissues and the presence of aging cells that influence homeostasis in joints. The SMS of senescent chondrocytes and resident MSCs contributes to the environmental imbalance by overproduction of chemokines, cytokines, and MMPs. These negatively affect cell susceptibility to growth factors and the ability to proliferate. As mentioned above, sMSCs express increased levels of the cell cycle inhibitor p16^INK4a. Using the mouse collagenase-induced OA model, Malaise et al. ⁸⁵ demonstrated that p16^INK4a-positive sMSCs (ie, resident articular osteochondral progenitors) might damage cartilage and participate in the progression of OA. In OA synovial tissue, protein p16^INK4a regulates the production of SMS-related factors that have a catabolic role^86,87. Elevated levels of SMS factor IL-6 have been observed in SF of OA and rheumatoid arthritis (RA) patients^88,89. Indeed, its long-term presence can prevent chondrogenic differentiation in SF-derived mesenchymal stem cells ⁹⁰ . Another line of supporting evidence was provided by Wei et al. ⁹¹ Authors reported that IL-6 suppressed the differentiation of murine BM-MSCs into chondrocytes in a dose-dependent manner. Furthermore, human knee OA chondrocytes show impaired proteasomal function, which is probably caused by reduced autophagic function ⁹² . Autophagy is a necessary process for maintaining homeostasis in healthy cartilage. Its deregulation leads to the accumulation of damaged macromolecules that can promote degenerative changes. The recently published study demonstrated that enhanced autophagy in chondrocytes delayed the progression of OA ⁹³ . The relationship between impaired autophagy and the increasing severity of OA has also been supported by other studies^94,95.

Repair and restoration of articular cartilage mainly depend on the proper function of chondrocytes and resident MSCs. S-MSCs are characterized by higher chondrogenic potential compared with other MSCs ⁹⁶ . Their cartilage rejuvenation properties rely on the production of hyaline cartilage matrices^86,97. However, a large number of senescent S-MSCs (sS-MSCs) may have a negative impact on these processes. It should be noted that sS-MSCs have a longer survival time than healthy S-MSCs due to their resistance to apoptosis. Accumulation of sS-MSCs in joints, cartilage, synovial tissue, and SF has been observed in many pathological conditions. S-MSCs are easily activated and able to proliferate in an inflammatory environment. Although, persistent inflammation in the OA joint can lead to their depletion and induce replicative senescence in these otherwise rapidly proliferating cells. Cao et al. ⁹⁸ showed that the synovium in the OA joint is the site with the high sS-MSC number. OA-related sS-MSCs displayed several characteristic features: cellular senescence, elevated pro-inflammatory molecule production, invasiveness, and reduced chondrogenic potential. However, the exact molecular mechanism by which resident sS-MSCs contribute to the disruption of homeostasis and loss of cartilage function is currently not well understood.

Of note, the number and size of extracellular vesicles (EVs) present in diseased/aged SF significantly differ from the young/healthy ones. Differences were also detected in the EV composition. Overall, the cargo of the young and old EVs primarily varies in the content of miRNAs, specifically the miR-183 cluster (miR-96/-182/-183) being strongly expressed in aged EVs. In the study conducted by Davis et al., mouse BM-MSCs were shown to easily engulf EVs isolated from the bone marrow interstitial fluid of old mice. These aged EVs suppressed the osteogenic differentiation in young BM-MSCs in vitro. Moreover, transfection of BM-MSCs with miR-183-5p decreased cell proliferation and osteogenic differentiation and favored senescence ⁹⁹ . Indeed, EV production is several times higher in senescent cells^100,101. Senescent EV-overloaded SFs also show an imbalance in levels of other miRNAs (eg, miR-199b, miR-185-5p, miR-661) ¹⁰² . Moreover, it was found that miR-34a in EVs derived from senescent chondrocytes can induce senescence in a paracrine way ¹⁰³ . Elevated levels of miR-31 in plasma have also been reported in patients with osteoporosis ¹⁰⁴ . Therefore, the in-depth analysis of EV cargo could be crucial for identifying senescent cells/environment and verifying diagnostic markers.

Rejuvenation

With increasing age, the extracellular environment is changing and affecting the fate of many cells. Pathological factors present in the microenvironment damage all cells, including resident MSCs. A better understanding of mechanisms involved in microenvironmental aging could contribute to its rejuvenation, maintaining the function of resident MSCs and preserving the healthy environment in many tissues at an older age. Several methods for sMSC rejuvenation are depicted in Fig. 2 and will be described in the following paragraphs.

Figure 2.

Techniques for the rejuvenation of senescent MSCs (sMSCs). This illustration shows various ways to maintain the stemness features in MSCs, possibly delaying/preventing the senescence-related adverse effects.

Rejuvenation of sMSCs by ECM, Antioxidants, Hypoxia, and Heat Shock

Generally, the niche is the local microenvironment in which MSCs naturally occur. It comprises ECM and bioactive molecules (growth factors, chemokines, cytokines, etc) ¹⁰⁵ . The ECM fills the space between the cells and provides support. Also, it affects cells’ behavior, growth, differentiation, and metabolism ¹⁰⁶ . Its exceptional composition allows MSCs to be kept in a quiescent state throughout life unless they are required to differentiate or repopulate damaged tissues. On the other hand, it can induce aging in MSCs, since the ECM composition changes with aging. Thus, through the modulation of ECM composition, the MSC biological activity may be retained until high cell passages. Lai et al. evidenced that the loss of BM-MSC characteristics was prevented when seeded on the human marrow stromal cell-derived ECM. This strongly enhanced cell proliferation and reduced the level of ROS. Furthermore, in vivo experiment confirmed that BM-MSCs grown on the ECM for multiple passages still retained the same ability for skeletogenesis. These findings suggest that large-scale BM-MSCs expanding on the marrow stromal cell-derived ECM is a promising technique for obtaining a high yield of highly functional BM-MSCs. Importantly, the establishment of a unique tissue-specific ECM can be utilized to control the MSC biological performance toward their desirable application ¹⁰⁷ . Currently, much attention is paid to HA, as it is the most prevalent glycosaminoglycan in cartilage and its level reduces with age ¹⁰⁸ . HA seems to have antiaging properties due to its ability to delay aging in murine AD-MSCs ¹⁰⁹ . Moreover, Wong et al. demonstrated that placenta-derived MSCs (PD-MSCs) treated with HA could preserve the expression of stemness markers (CD105, CD90, and CD73) and osteogenic potential for 19 passages when compared with PD-MSCs cultured only on normal cell culture surface. The long-term HA treatment maintained the replicative capacity of PD-MSCs even after their transfer to the standard cell culture surface. Authors suggest that HA-related positive effects may be executed via the regulation of cytoskeleton protein distribution ¹¹⁰ . HA also protects MSCs from DNA damage caused by oxidative stress. And presumably, it is responsible for the decreased ROS formation in H₂O₂-treated PD-MSCs.

Further, in vitro experiments have shown that exposure of BM-MSCs to 3% hydrogen leads to an increased number of cell cycles without any loss in differentiation potential and paracrine activity. However, 3% hydrogen gas treatment did not reduce hydroxyl radical, protein carbonyl, and 8-hydroxydeoxyguanosine, indicating that scavenging hydroxyl radical might not be the primary mechanism behind the antiaging effect. In addition, hydrogen exposure is linked to the delay of senescence, thanks to decreased expression of SA-β-gal and p16^INK4a, and altered paracrine activity in BM-MSCs. Even though hydrogen treatment did not reduce the paracrine activity of BM-MSCs, it changed the level of secreted components. For instance, increased secretion of basic FGF, HGF, and indoleamine 2,3-dioxygenase was detected ¹¹¹ . Other antioxidants, such as ascorbic acid (AA) ¹¹² , Cirsium setidens ¹¹³ , and lactoferrin ¹¹⁴ were confirmed to suppress ROS production and reduce the signs of aging in MSCs as well.

Notably, a hypoxic environment also shows a significant antisenescence effect on MSCs. Hypoxia-exposed MSCs can preserve stemness characteristics without increasing the risk of tumorigenicity ¹¹⁵ . So far, its beneficial effects have been demonstrated on MSCs derived from different tissues. For example, human amniotic fluid mesenchymal stem cells (AF-MSCs) cultured at low oxygen concentration (1% O₂) preserved features of stemness, proliferation, and osteogenic potential. Hypoxic AF-MSCs underwent a metabolic shift and showed increased resistance to pro-apoptotic stimuli ¹¹⁶ . Under hypoxic conditions, AD-MSCs displayed more efficient cell proliferation with a faster population doubling rate and enhanced secretion of multiple angiogenic growth factors. And hypoxia-inducible factor (HIF) is considered the main executor of hypoxic response ¹¹⁷ . Similarly, a study by Sheng et al. proved the stimulatory effect of 2% hypoxia on proliferation, differentiation into endothelial cells, and vascular endothelial growth factor (VEGF) expression in BM-MSCs. The PI3K/Akt pathway was shown to have an essential role in this hypoxia-triggered boost ¹¹⁸ . Moreover, in mouse BM-MSCs, hypoxia (3% O₂) contributes to the evaluated expression of chemokine receptors CXCR4 and CXCR7 implicated in cell migration and survival in vitro. Hypoxia preconditioning not only facilitated MSC chemotaxis and viability but also promoted the release of proangiogenic and mitogenic factors. The increased homing of hypoxia-exposed BM-MSCs was related to higher functional recovery, enhanced mitogenic response, and decreased apoptotic cell death in vivo ¹¹⁹ . In addition, exposure of olfactory mucosa-derived mesenchymal stem cells (OM-MSCs) to 3% hypoxia delayed senescence, enhanced survival after transplantation, and improved the neuroprotective effects in an intracerebral hemorrhage mouse model. Hypoxia ameliorated the function of OM-MSCs by upregulating the miR-326/PTBP1/PI3K-mediated autophagy ¹²⁰ .

A prolonged exposure to heat shock (HS) can be harmful and trigger premature cellular senescence in human MSCs derived from menstrual blood (MB-hMSCs). HS-treated MB-hMSCs exhibited altered morphology (enlarged and flattened shape) as well as other senescence-related features, such as increased SA-β-gal activity and expression of cell cycle inhibitors ¹²¹ . However, other studies showed that HS can have also positive effect and improve the biological performance of aging MSCs. Exposure of AD-MSCs to HS (41°C for 60 min) once in a week led to increased expansion and differentiation potential, particularly at higher passages (eight passages cultured for approximately 7 weeks). Stressed AD-MSCs had elevated levels of SIRT-1, which might be related to the delayed senescence in these cells ¹²² . Similarly, in BM-MSCs, Wang et al. evidenced that HS pretreatment (42°C for 60 min) inhibited the apoptosis and enhanced the survival under cisplatin-induced chemotherapy environment. Most likely, this effect was associated with the increased expression of HSP70 and HSP90 ¹²³ . The antiapoptotic effect of HSPs may be executed via binding the caspase-recruitment domain of apoptotic protease-activating factor-1 and the caspase-independent death effector apoptosis-inducing factor leading to the suppression of the apoptotic pathway^123,124. Interestingly, addition of exogenous HSP70 supported growth of aged but not young murine AD-MSCs. The stimulatory effect was observed also after application of a mild HS (42°C for 5 min). Importantly, aged murine AD-MSCs were significantly more responsive to higher heat stress compared with the young cells ¹²⁵ .

Rejuvenation by Soluble Factors

Multicellular organisms do not have the ability and are not equipped with mechanisms to avoid the natural aging process. The best-known technique used to study vertebrate aging is heterochronic parabiosis, which represents the surgical connection of the circulatory system of old and young mice. This technique was discovered in 1864 by physiologist Paul Bert and later used to elucidate the improved stem cell aging in old mice after exposure to the young mouse circulatory system. It was found that factors in young blood contributed to the activation of biochemical pathways in aged stem cells and boosted their regenerative properties ¹²⁶ . The ability of blood derivatives to affect the regeneration of stem cells and tissues is also consistent with the findings of several other studies^127,128. As an example, platelet-rich fibrin and platelet-rich plasma include a wide range of mediators [eg, fibroblast growth factor 2 (FGF-2), IGF-1, platelet-derived growth factor (PDGF), and VEGF] that can promote MSC proliferation, differentiation, and maintenance of stemness characteristics^129,130. The rejuvenation of long-term cultured senescent human bone–derived MSCs by human platelet lysate enhanced cell phenotype and proliferative activity ¹²⁷ . Further research also confirmed the interaction of circulatory factors with signaling pathways involved in the activation or inhibition of tissue-specific MSCs. The reduced activity of MSCs in old mice is probably related to the dominance of inhibitory factors circulating in their blood. For instance, TGF-β is increased in the plasma of aged mice ¹³¹ . Gurung et al. demonstrated that TGF-β receptor (TGF-βR) inhibitor promoted endometrial mesenchymal stem cells (eMSCs) proliferation in the undifferentiation state during prolonged in vitro cultivation. Suppression of TGF-βR signaling further increased cell proliferation, and prevented apoptosis and senescence in eMSCs ¹³² . It is plausible that due to the high level of circulating TGF-β in the blood of old mice, aged subjects have a reduced ability to regenerate damaged tissue compared with young ones^133,134.

Several lines of evidence indicate that all molecular and phenotypic features of aging are reversible and thus can be rescued by applying soluble factors and cells from a young circulating environment ¹³⁵ . The overexpression or supplementation of systemic growth differentiation factor 11 (GDF-11), which normally decreases with age, improved homing of endothelial progenitor cells and angiogenesis in old ischemic hearts, enhanced muscle structural and functional features, and supported strength and endurance exercise capacity in aged mice^131,136. Nevertheless, some studies have reported contradicting results regarding the positive effect of GDF-11, especially on skeletal muscle stem cells^137,138. Delayed senescence of BM-MSCs was also achieved by stimulation with factors such as AA, epidermal growth factor (EGF), FGF-2, and PDGF. Although BM-MSCs exhibited an elevated in vitro expansion, their differentiation potential was reduced before reaching senescence. The expression of stem cell surface markers was not affected by the loss of differentiation capacity ¹³⁹ . In general, growth factors, such as FGF-2, show pleiotropic effects, and their application may generate ambiguous data. Several studies have demonstrated a dual effect of FGF-2, both stimulatory and inhibitory, on the differentiation of MSCs. Hanada et al. ¹⁴⁰ showed that FGF-2 markedly augmented cell growth and triggered osteoblastic differentiation in rat BM-MSCs. Moreover, its positive effects on chondrogenic differentiation in human BM-MSCs have also been reported ¹⁴¹ . On the other hand, the study by Baddoo et al. demonstrated an inhibitory effect of FGF-2 on multilineage differentiation of mouse BM-MSCs. Its removal from culture restored the differentiation potential of BM-MSCs. Therefore, FGF-2 may be used selectively when the expansion of undifferentiated BM-MSCs is required ¹⁴² . The study by Coutu et al. evidenced that murine BM-MSCs cultured without FGF-2 exhibited distinct features of cellular senescence at very early passages. Interestingly, the senescent phenotype of BM-MSCs expanded without FGF-2 cannot be reversed by the subsequent stimulation with FGF-2. To further clarify the effect of FGF-2, cells cultured in a medium supplemented with FGF-2 for three passages were subsequently grown in its absence. FGF-2 removal immediately triggered growth arrest in BM-MSCs ¹⁴³ . Although numerous factors were shown to rejuvenate MSCs, their mechanism of action is still unknown. It is hypothesized that these youth-stimulating factors may operate nondirectly. Their activity possibly depends on epigenetic mechanisms or interaction with other components ¹⁴⁴ .

Rejuvenation by the Secretome of the Young MSCs

MSCs secrete various types of EVs: exosomes (40–150 nm), microvesicles (150–1,000 nm), apoptotic bodies (50–2,000 nm), and others. EVs are frequently enriched with bioactive molecules (serum proteins, angiogenic and growth factors, hormones, cytokines, ECM, etc). The composition of EVs and secretome affects the condition of niche and resident cells. It also plays a significant role in protection against various diseases, especially if it comes from young and healthy cells ¹⁴⁵ . Indeed, young secretome has a noticeable rejuvenating effect. On the other hand, aging cells typically release harmful cargos that can cause inflammation, epigenetic changes, dysfunction of cell organelles, and cellular senescence. Old MSCs cultured with ECM or a conditioned medium from young MSCs showed improved differentiation and replication ability ¹⁴⁶ . The study performed by Liang et al. demonstrated the positive impact of human UC-MSC secretome on the aging rat BM-MSCs. Human UC-MSC secretome loaded into silk fibroin-based hydrogels was proved to recover MSC potential and attenuate the local bone loss in vivo ¹⁴⁷ . In addition, a cocktail of molecules, including EGF, FGF, VEGF, and IGF from the secretome of UC-MSCs, may prevent the onset of osteoporosis, thanks to bone remodeling and regeneration^148,149. Several studies have demonstrated a significant rejuvenating potential of molecules that are part of EV cargo derived from young cells on aging human MSCs^150–152. More specifically, after EV-containing nicotinamide phosphoribosyltransferase (NAMPT) internalization into the cell, NAD⁺ biosynthesis was elevated. Furthermore, supplementing extracellular NAMPT through EVs obtained from young mice markedly augments the wheel-running activity and expands the lifespan of aged mice suggesting a potential antiaging treatment in humans ¹⁵³ . The favorable effect of secretome is supported also by the study conducted by Arasu et al. Authors showed that BM-MSCs secreted HA-coated EVs that carried mRNAs for CD44 and all HAS isoforms. This indicates that HA-coated EVs may be used mainly when tissue regeneration requires HA supplementation, as in the case of aging articular cartilage ¹⁵⁴ . Currently, considerable attention is paid to EVs derived from urine-isolated stem cells (USCs-EVs) since these cells are easily available and do not require invasive collection methods. USCs-EVs carry several components in abundance, including collagen triple-helix repeat containing 1 and osteoprotegerin. These mitigate bone loss by supporting osteoblast and suppressing osteoclast activity in osteoporotic mice. Importantly, USC-EV properties were not affected by age, gender, or health condition of the USC donor. Data clearly indicate the medicinal potential of autologous USCs-EVs as well as show a new promising strategy in osteoporosis treatment ¹⁵⁵ . A remarkable rejuvenating potential of antler stem cells (ASCs)-derived exosomes (ASCs-EVs) on human sMSCs was recently demonstrated. Generally, ASCs show high proliferative and regenerative capacity since they annually initiate the formation of the entire organ. After ASCs-EVs treatment, human MSCs showed attenuated signs of senescence. Also, intra-articular injection of ASCs-EVs reduced cartilage damage in the OA mouse model. Taken together, ASCs could serve as an inexhaustible source of EVs to support cell-free therapy ¹⁵⁶ . The effect of EVs harvested from hypoxic nonsenescent dental pulp stem cells (DPSCs) on normoxic prematurely senescent DPSCs was also investigated. Upon EV treatment, normoxic senescent DPSCs exhibited suppressed features of senescence, promoted expression of stemness markers, and metabolic switch toward glycolysis. EV components increased miR-302b and HIF-1α levels in acceptor cells. Notably, miR-302b was also encapsulated in EVs derived from hypoxic nonsenescent DPSCs. It was proposed that the exogenous and endogenous upregulation of miR-302b may induce HIF-1α. And this pathway is likely behind the delaying DPSC aging ¹⁵² . Indeed, the properties of MSC-isolated EVs may be modulated by environmental factors, such as low oxygen tension or hydrodynamic culture in bioreactors. 3D aggregation of BM-MSCs significantly increased the quantity (by twofold) and miR-21 and miR-22 expression, and changed the protein composition (eg, upregulation of cytokines and anti-inflammatory factors) of EVs compared with 2D culture. 3D BM-MSC-EVs rejuvenated sMSCs and displayed promoted immunomodulatory capacity ¹⁵⁷ . Application of EVs represents a relatively safe cell-free approach to rejuvenate senescent cells since it does not induce tumor formation and shows low immunogenicity. Despite having many advantages, several drawbacks, such as short lifespan, degradation, and inefficient targeting, significantly reduce the effectiveness of EV-based techniques. The production of artificial EVs could, at least partially, eliminate these shortcomings.

Rejuvenating Approaches in Animal Models

Even if the rejuvenating techniques have not reached the clinical phase yet, their application in in vivo models is established. To show that these approaches do have a potential to enter clinical research we provide data from several preclinical studies.

Exosomes derived from young rat BM-MSCs (BM-MSCs-Exos) were locally administered and their impact on bone regeneration was investigated in older rats. Significant acceleration in healing process and better mechanical properties were demonstrated upon BM-MSCs-Exos treatment ¹⁵⁸ . EVs secreted from hypoxia-pretreated BM-MSCs (5% O₂ for 48 h) exerted a remarkable therapeutic outcome in the rat OA model. The chondroprotective effect of hypoxic EVs may be mediated by the miRNA-18-3P/JAK-STAT or miRNA-181c-5p/MAPK signaling pathway, which might enhance chondrocyte proliferation and migration and inhibit chondrocyte apoptosis ¹⁵⁹ .

Moreover, an increased anti-inflammatory and regulatory potential of allogeneic BM-MSCs primed with tumor necrosis factor-α and IFN-γ was reported in an equine model of chemically induced OA ¹⁶⁰ . Recently, an innovative approach has been brought by Pei et al. Cell-derived decellularized extracellular matrix (C-dECM) rejuvenation improved the infrapatellar fat pad–derived stem cell (IPFSC) cartilage engineering and functional regeneration in a rabbit model of osteochondral defect. The profound effect was observed particularly for the decellularized ECM deposited by urine-derived stem cells (UdSCs). Authors suggested that mesenchymal–epithelial transition and inflammation-mediated macrophage activation and polarization are presumably implicated in the C-dECM–related enhancement of IPFSCs’ chondrogenic potential ¹⁶¹ .

Chromobox protein homologue 4 (CBX4) is downregulated in aged human MSCs. The CBX4 deficiency causes destabilized nucleolar heterochromatin, increased ribosome biogenesis, elevated protein translation, and premature aging. CBX4-overexpressing lentiviral vectors delivered by intra-articular injection dampened cartilage erosion, improved bone density, triggered the expression of bone growth and differentiation genes, and downregulated the expression of aging and inflammation molecules in the OA model ¹⁶² . In addition, yes-associated protein (YAP) is a well-known effector of Hippo signaling, which represses the senescence of human MSCs via activation of forkhead box D1 (FOXD1) transcription. The intra-articular delivery of lentiviruses encoding YAP or FOXD1 decreased the number of senescent cells and suppressed articular inflammation and cartilage erosion. Thus, gene therapy via the introduction of geroprotective factors leading to rejuvenating sMSCs may represent a new tool in the OA therapy in the future ¹⁶³ .

Conclusion

MSC senescence is integral to the human aging process and is one of the underlying causes of joint-related diseases such as OA. In vitro and in vivo MSC senescence is a major obstacle, especially the development of SMS, which not only changes the MSC biological activity but also affects other resident cells and their microenvironment. Mechanisms involved in regulating MSC senescent phenotype and the pathophysiological impact of sMSC-mediated autocrine and paracrine signaling are still not completely understood. On the other hand, parabiosis has suggested over decades that factors derived from young blood may support the regeneration of diseased or aged tissues. Currently, many rejuvenation strategies (from cell free to gene therapy) used for delaying and/or reversing cellular senescence are being scrutinized. These can be a very powerful tool for maintaining the proper function of resident MSCs and boosting the therapeutical potential of autologous MSCs isolated from older people before transplantation. MSCs represent an essential source for cellular and cell-free therapies in managing many disorders, including degenerative and other aging-related diseases. Therefore, a better understanding of senescence itself enables control of the course of MSC-based treatments to maximize clinical benefits.