Age-related Changes in Bone Marrow Mesenchymal Stromal Cells

Introduction

Aging is a gradual process marked by the deterioration of functionality in living organisms with the passage of time. It is a complex phenomenon known to be affected by a variety of factors like diet, lifestyle, environment, heredity, and disease. It slows down the biological mechanisms that aid tissue maintenance, immunity, and health. Disorders like osteoarthritis (OA), osteoporosis (OP), Alzheimer's, and Parkinson's diseases have been associated with aging, making the life of patients more challenging as compared to a healthy elderly population. Physically, locomotion and daily activities become difficult, and vulnerability toward infections increases with aging.

OP is a skeletal disease characterized by a reduction in bone mineral density, predisposing people to an increased risk of fracture. The hypothesis that OP is an age-associated disease and mainly a consequence of estrogen deficiency was proposed in 1941. ¹ In 1998, the “unitary model of OP in postmenopausal women and aging men” was put forward, this concept combined postmenopausal (involving mainly trabecular bone) and senile OP affecting both cortical and trabecular bone. ² The management of OP involves various pharmacological options that can be divided into antiresorptive and bone stimulatory agents (the former are more broadly used in clinical practice). On the other hand, OA is a degenerative joint disease also associated with age that includes a group of pathologies of joint structures resulting in pain and disability. There is no disease-modifying treatment for OA, and the management of patients is currently limited to pain reduction and lifestyle modification. Increase in age is one of the risk factors for OA. ³ However, it is now recognized that aging and OA are independent processes, so that age is one of many contributory factors. Because OA has been historically described as a disease of the cartilage, most of the research on aging in OA has been focused on cartilage tissue. ⁴ More recently, OA has been proposed as a disease of the whole joint, which is additionally characterized by severe alterations to subchondral bone as well as by low-grade systemic and local inflammation. ⁵ Abnormalities in OA bone include subchondral plate sclerosis, bone marrow (BM) lesions, and the formation of osteophytes, all are a result of intermittent abnormal subchondral bone remodeling, which links this aspect of OA to the OP. ⁶

One of the key cellular players in bone physiology is the osteoblast, a bone-forming cell derived from BM resident mesenchymal stromal cells (MSCs). BM-MSCs are self-renewing in vivo, and the shift in their differentiation from osteogenic to adipogenic lineage with increase in age ^{7 –10} can potentially affect osteoblast formation and bone remodeling, in general, and thus be a factor in OP and OA development and pathogenesis. In vitro expanded MSCs have broadly been defined by their adherence to plastic, ability to give rise to at least 3 cell lineages (adipogenic, osteogenic, and chondrogenic); presence of CD73, CD90, and CD105 surface molecules; and absence of CD11b or CD14, CD19 or CD79a, CD34, CD45, and human leukocyte antigen—antigen D related (human leukocyte antigen—antigen D related [HLA-DR])—as defined by the International Society for Cellular Therapy. ¹¹ However, the markers used for characterizing in vitro MSCs are not very useful for the identification of MSCs or their subpopulations in vivo. ¹² Most of the experiments for understanding MSC aging have been performed using cultured MSCs, and attempts have been made to correlate the findings with in vivo environment. This article addresses the comparison of in vitro MSC aging with in vivo MSC aging and discusses the concept of in vivo MSC “rejuvenation” and its future prospects for novel therapies for OP and OA.

General Theories of Aging

In an attempt to understand MSC aging, it is worth reflecting on the general theories of cellular and whole body aging. Among many existing theories, DNA damage, the free radical (FR), telomere shortening, and stem cell theories have been discussed in detail recently. ¹³

The DNA damage theory is among the earliest theories focusing on the mechanism of aging. It refers to the accumulation of DNA damage in every cell with the passage of time. ^14,15 The sources of DNA damage vary from endogenous toxic chemicals and reactive oxidative species (ROS) to exogenous ultraviolet light. This theory has been explored to describe in vitro MSC aging, particularly in relation to the accumulation of chromosomal instability following prolonged culture. ¹⁶

The FR theory or the theory of ROS proposes that the continuous process of FR generation in metabolism causes damages to tissues owing to the presence of free and unpaired electron on the oxygen atomic outer shell. ¹⁷ Antioxidants in the body combat the negative effects of ROS. However, when the balance between production of antioxidants and FR is lost, oxidative stress (OS) develops. ¹⁸ This can then lead to oxidation of nucleic acids, severely damaging the DNA and resulting in mutagenesis and modification of the transcription of specific genes. In terms of MSC aging, this theory has primarily been applied to describe the decrease in adhesion of MSCs ¹⁹ and the increased bias toward adipogenic differentiation. ²⁰

The telomere shortening theory postulates that with each cell division, the length of telomeres in mature cells keeps reducing due to the attrition and the lack of telomerase activity, an enzyme responsible for telomere elongation. In BM-MSCs, telomere attrition can also occur leading to cellular senescence, ²¹ although its exact mechanism remains unclear, as in MSCs, telomerase activity is present but low. While telomere length decreases with an individual’s age, recent data suggest that after the age of 75, it becomes positively correlated with age implying its significance for survival in the very old age. ^15,22 In terms of MSC aging, this theory has been extensively tested using serially passaged MSCs, where an average 17 base pair telomere loss with each population doubling was confirmed. ^21,23

The stem cell theory of aging postulates a decline in stem cell number and functionalities as a potential effect of aging. ^{24 –26} Schultz and Sinclair have discussed that the probable causes of cellular aging (i.e., the abovementioned telomere attrition, DNA damage, and cellular senescence) can also be applied to stem cells. ¹⁵ Fukada et al. have not only described the aging of stem cells but have also connected the other theories with the stem cell theory to explain the complex progression of aging. ¹⁸ As BM-MSCs can be viewed as true stem cells, ²⁷ the stem cell theory is linked to each of the theories mentioned above.

Finally, the theories of epigenetic alterations ²⁴ and decline in protein homeostasis ²⁵ have also gained importance in the recent past. While all these theories have been mentioned separately, they are interlinked and lead to one another. Mitochondrial damage often leads to the production of ROS that interact with molecules in the body affecting them negatively. ²⁴ The accumulation of the oxidative by-products of ROS interactions, with time, causes DNA damage and genomic instability resulting in cellular senescence. This disturbs the self-renewal capacity of stem cells in their microenvironment which is known to interrupt the normal functioning of different types of stem and other cells in the BM. ^18,28

In Vitro MSC Aging

This section outlines the recent findings of in vitro MSC aging as well as the current methods used to track MSC aging in culture.

Passage Dependent

It is well accepted that MSC expansion in culture results in their accelerated aging. ^23,29 Colony-forming unit fibroblast assay has been one of the oldest methods to document the loss of proliferation in cultured MSCs during their extended passaging. ³⁰ Many independent studies have later documented that this loss of proliferation is correlated with a decline in the telomere length with increasing population doublings (PD). ^23,29 The senescence-associated β-galactosidase (SA-β-gal) enzyme activity is increased in senescent cells and similarly, the number of SA-β-gal positive cells have been reported to escalate with increase in PDs. ^29,31 Regarding the potentials of MSC for differentiation, a study by Vacanti et al. was among the first to show a significant decrease in the differentiation potential of cultured BM-MSCs (porcine) into osteogenic lineage in the late passages as compared to early passages. The late passages also exhibited actin accumulation, reduced adherence to substrates, and increased activity of β-galactosidase all of which are indicative of cellular aging. ³² More recently, Yao et al. have highlighted the effects of in vitro aging on wound healing ability of MSCs derived from mouse fat pads. ³³

In addition to the abovementioned changes, Bonab et al. observed anomalies in the morphology of human BM-MSCs with increasing passage number and suggested that in vitro aging of MSCs begins from the minute they are plated on plastic. ²³ One of the potential reasons for such accelerated aging could be the fact that MSCs placed in culture receive conditions supportive of their enhanced proliferation that goes far beyond its physiological demands (Fig. 1). Importantly, MSCs are cultured under hyperoxia, where they consume oxygen at a high rate, ³⁴ leading to high levels of ROS that accumulate with prolonged culture of MSCs. ³⁵

OPEN IN VIEWER

Figure 1.

A proposed model highlighting the different mechanisms of bone marrow–mesenchymal stromal cells (MSCs) aging in vitro and in vivo. 2-D = 2-dimensional; 3-D = 3-dimensional; ECM = extracellular matrix; HSC = hematopoietic stem cells.

One of the suggested solutions to such accelerated aging could involve growing MSCs under hypoxic conditions. Under long-term exposure to hypoxia, MSCs may adapt themselves to low oxygen levels, decreasing their oxygen consumption and reducing their ROS production. ²³ This would lead to improved MSC survival and reduced apoptosis. Interestingly, the migration of adipose tissue MSCs may also be induced by hypoxia as shown in animals. ³⁶

In Vivo MSC Aging

In vivo tracking of MSC aging has so far been performed only in animal models. While the available literature broadly suggests a decline in MSC frequency with aging, ^47,48 the issue remains controversial. For example, some studies have indicated a decline in MSC number in older individuals, ^31,49 whereas other scientists did not find any significant changes. ^43,50 This could be due to different volumes of BM aspirate used as well as different processing methodologies (e.g., direct plating vs. density centrifugation). ^51,52 One of the most interesting features of MSCs from older individuals appears to be their reduced propensity for osteogenic differentiation with increased bias toward adipogenic differentiation. The loss of balance between osteogenic–adipogenic differentiation leads to increased BM adiposity and is seen in OP. ^7,10,28,53 The exact mechanism underlying the adipogenic bias is yet to be clearly understood. Recent studies have identified microRNAs miR-27a, ⁵⁴ miR-27b, Let-7G, and miR-106a ⁵⁵ that are necessary for maintaining the osteogenic differentiation potential of MSCs and have displayed a significant decline with aging.

As mentioned in the above sections, the great majority of studies investigating the aging of human BM-MSCs have been performed on MSCs expanded in culture. As these cultured MSCs undergo an aging process as a result of extensive proliferation during consecutive PDs, these studies provide only indirect evidence of how MSCs may age in vivo. Furthermore, in vivo BM-MSC niche is not mirrored when these cells are cultured and expanded on the plastic surfaces. Thus, it can be proposed that in vivo and in vitro aging of MSCs may have both common (overlapping) and nonoverlapping features (Fig. 1).

Rejuvenation of OP and OA MSCs

Early studies have shown a decline in the number of osteoblast progenitors in the BM of OP patients compared with age-matched controls. ⁸⁰ At least in females, this could be explained by an altered balance in the systemic levels of hormones such as estrogen and testosterone. Culture-expanded MSCs from OP patients have been shown to possess lower proliferative and osteogenic capacities ⁸¹ (i.e., to exhibit distinctive marks of premature in vitro aging) compared to healthy individuals. Recent evidence suggests that this could be due to the overexpression of osteogenic inhibitors in OP MSCs. ⁸² A recent study by Zhou et al. has identified a number of differentially expressed genes that are up- or downregulated in OP MSCs, which could not only be used as a biomarker for OP, ⁸³ but also serve as potential targets for therapeutic modulation to rejuvenate OP MSCs. MSCs extracted from the areas of subchondral bone damage in hip OA patients also appear to acquire several abnormalities indicative of premature aging (e.g., reduced in vitro proliferative and mineralization capacities). ⁸⁴ It remains to be investigated whether putative accelerated aging of OP and OA MSCs in vitro is reflective of their accelerated aging in vivo. However, as mentioned in the above sections, OP MSCs may indeed reside in a considerably modified in vivo niche that is markedly enriched in marrow fat. ⁵³ Changes in marrow adiposity and an increased cellular necrosis have also been detected in the affected areas of OA subchondral bone. ⁸⁵ This suggests that targeting putative in vivo aging in OP and OA MSCs may be performed indirectly, by targeting their in vivo niches (Table 1).

OPEN IN VIEWER

Table 1.

Abnormalities in MSCs or Their In Vivo Niches as Potential Targets for OP and OA MSC Rejuvenation.

Abnormality	Intrinsic/Environmental	OP	OA
MSC number	Intrinsic	Reduced 80	Unbalanced in damaged areas 84,86
Signaling pathways in MSC	Intrinsic	Reduced osteogenesis due to overexpression of osteogenic inhibitors 82	Abnormal homing due to altered TGFβ signaling 86 and chemokine receptor expression 84
MSC niche: ECM	Environmental	Increased BM adiposity 53	Increased BM adiposity 85 Increased levels of free TGFβ 86
MSC niche: Neighboring cells	Environmental	Accelerated osteoclast activation 5 3	Increased numbers of osteoclasts in damaged areas 86

Note. BM = bone marrow; ECM = extracellular matrix; MSC = mesenchymal stromal cells; OA = osteoarthritis; OP = osteoporosis; TGFβ = transforming growth factor β.

Abdallah et al. tested the impact of serum from young (20–30 years old) and old (70–84 years old) donors on MSC differentiation and proliferation and found a significant decline in osteogenic differentiation potential of MSCs grown in serum from old donors. ⁸⁷ This suggests the existence of, yet unknown, factors in the serum or plasma from “younger” donors that could potentially be used systemically to reduce the rate of in vivo aging of MSCs, including OP MSCs. Similarly, Sun et al. cultured MSCs from young and old mice on plastic, young extracellular matrix (ECM) and old ECM and have shown that the number and quality of MSCs from old mice could be improved when cultured on young ECM. ⁸⁸ In vivo ECM modulation could thus be another strategy to rescue or rejuvenate MSCs in OP and possibly, OA.

The “mechanistic target of rapamycin” (mTOR) is a serine/threonine protein kinase of the phosphatidylinositol-3-OH kinase (PI3K) family, which regulates cell growth, metabolism, and functions in two complexes—mTOR complex 1 and mTOR complex 2. This pathway has been the target of wide interest ever since it was discovered, and it has been suggested that inhibition of this pathway could extend the life span of rodents. ⁸⁹ The role of rapamycin in the differentiation of MSCs or their progenitors has also been studied, but to date, the results are controversial, ⁹⁰ suggesting that the drug can either encourage ⁹¹ or discourage ⁹² osteogenic differentiation of MSCs. This implies the need for further exploring this pathway and related molecules to understand their function in MSC aging and their possibility to rejuvenate aged MSCs.

Sirtuins (SIRT) genes have similarly proved to be of significant importance in regulating aging. ⁹³ Simic et al. performed experiments on mice with MSC specific sirtuin 1 (Sirt-1) knock out gene and found that Sirt-1 regulates the MSC differentiation by the deacetylation of β-catenin. ⁹⁴ This suggests that targeting this gene using gene therapy can be used for halting or reversing aging of MSCs. However, further studies are required to understand the connecting links between the SIRT genes and this signaling cascade to ultimately apply this approach for preventing aging of MSCs in vivo and in diseases such as OP and OA.

Conclusions and Future Directions

Currently, our understanding of MSC aging in vitro considerably surpasses our understanding of MSC aging in vivo. In order to move forward in the study of human MSC aging in vivo, an ability to isolate a pure population of uncultured MSCs represents an initial and critical step. Up to now, a broad consensus on the combination of markers to purify human BM-MSCs does not exist, although markers such as CD271 and MSCA-1, ¹² possibly in combination with CD140a, ²⁷ appear to be the most promising.

Methods currently used for the detection of aging in cultured MSCs could, in principle, be used for the study of aging of uncultured MSCs, although these need to be adapted to be used with very low numbers of cells as BM-MSCs are very rare in vivo. Research based on molecules like prelamin and lipofuscin, which can be ⁶⁵ detected at the later stages of MSC culture in vitro, is a promising approach for identifying “aged” and senescent MSCs in vivo. ^39,95 It will be interesting to investigate the change of size of BM-MSCs with aging in vivo and compare it to the acknowledged change of size in vitro ⁴⁰ to possibly use cell size as an indicator of MSC aging in vivo. Churchman et al. have demonstrated a decline in the expression of molecules like connexin 43 in uncultured human BM-MSCs in older donors using quantitative polymerase chain reaction, ⁹⁶ suggesting that it, and similar surface molecules, could be considered further as potential indicators of aged MSCs in vivo.

The ability to identify MSCs in situ and extract them less invasively from their native niches would then lead to a better understanding of their local environments, which could then be reconstructed and modeled using 3-dimensional (3-D) “organ” cultures. These organoids could then be used for developing and screening new rejuvenation compounds to target OP and OA MSCs in 3-D environments most closely resembling their native niches.

These approaches should lead to a better understanding of MSC aging in vivo. Considering that MSCs form an integral part of the musculoskeletal system and that OP and OA have been associated with aging, ^1,3 prevention of MSC aging in vivo could lead to novel therapies to target altered bone formation in OP and OA.