Aging of bone marrow mesenchymal stromal/stem cells: Implications on autologous regenerative medicine

1. Introduction

Many studies concerning MSC aging were published in the last years. In these studies comparing cells from young and old donors, various parameters are analyzed like proliferation, clonogenicity, differentiation potential, immunomodulation properties, gene expression, migration and cellular adhesion [1]. Unfortunately the studies do not always present the same conclusions and that can be explained by the multitude of experimental variables of the different works [2]. In addition, it is difficult to conclude if replicative aging of MSC reflects its in vivo ageing. This is why it is important to take into account the passage of the cells used for in vitro experiments to compare the studies and to work with cells in the early passage to approach as much as possible the conditions of the MSC in vivo. Nevertheless, because regenerative medicine requires amplification of the cells (for several passages) in order to have a sufficient number of cells for the clinical trials, it is thus very important to study replicative aging in working with late passages cells.

2. Bone-marrow mesenchymal stromal cells aging and cell proliferation

Most of the studies published reported a decrease in the in vitro proliferation capacities of human bone marrow MSC (hBM-MSC) with the increase in the age of the donors [2–10]. In addition, studies carried out by Siegel et al. showed that hBM-MSC coming from female donors exhibited higher proliferation capacity compared to hBM-MSC coming from male donors [11]. However, some published studies did not find differences in proliferation rates between hBM-MSC coming from donors of different ages. That can be explained by the proximity of the ages of donors taken into account for the analysis compared to the majority of the previous studies.

Clinical use of MSC require an in vitro expansion step, where they undergo a replicative aging process [12]. This phenomena is characterized in particular by the entry of cells in senescence induced by the shortening of the telomeres [13]. Indeed, reduction in the in vitro proliferation capacity of hBM-MSC was correlated with the shortening of telomeres [3] and with the increase in senescence [14].

3. Bone-marrow mesenchymal stromal cells aging and clonogenicity

Among the first studies on the clonogenicity of the hBM-MSC, the work of Digirolamo et al. showed that the capacity of hBM-MSC to form colonies is correlated with the capacity of proliferation [15]. Moreover, the work of Sekiya et al. showed that it is necessary to obtain a compromise under the seeding conditions of hBM-MSC at the moment of cell expansion, in order to have a good level of cell amplification and clonogenicity [16–18].

Moreover, Digimolamo et al. showed that there exists a link between the size of the colonies and the multipotence of hBM-MSC [15]. Indeed, they showed that big colonies are more easily differentiated into osteocytes and adipocytes, but that the capacity of adipocyte differentiation disappears with the in vitro expansion independently of the age of the donors.

Most of the studies showed a decrease in the number of CFU-F with the increase of the age of hBM-MSC donors [2–4,8,19,20]. However, there exists a number much less significant of studies which showed than the age of hBM-MSC donors does not have any influence on the number of colonies [21,22] and that the hBM-MSC coming from young female donors express more colonies compared to old female donors [11].

4. Bone-marrow mesenchymal stromal cells aging and phenotype

The clusters of differentiation (CD) present at the surface of the hBM-MSC highlighted in various studies are in agreement with the ISCT (International Society for Cellular Therapy) [23]. However, no obvious trend concerning the phenotypical markers and the age of the donors was observed [2]. Moreover, cells undergo several of in vitro passages which are known to deteriorate the expression of cellular surface antigens [2]. This is why, some authors suggest studying the subpopulations of hBM-MSC starting from total marrow or early in vitro passage [11]. Indeed, Siegel et al. show that even in absence of correlation between the age of the donors and the expression of the surface markers on hBM-MSC, there would be a subpopulation of cells expression CD146 at the first passage having a smaller size and better proliferative capacities [11]. The authors also showed that the hBM-MSC coming from young donors contain more positive cells for the markers CD71, CD146 and CD274. Other authors showed that this population of hBM-MSC CD146+ decreases with the donors age [24]. The expression of the marker CD146 in MSC would be even correlated with their multipotence [11,20].

5. Bone-marrow mesenchymal stromal cells aging and multipotence

Most of the studies showed that adipogenic, osteogenic and chondrogenic differentiation of hBM-MSC is independent of the age of the donors [2,7,11,25,26]. Some authors showed that the capacity of differentiation of hBM-MSC would be rather correlated with the capacity of proliferation at the time of the induction of the differentiation and not with the age of donors. Indeed, more the cells have a high proliferation capacity at the time of the induction of differentiation, more the differentiation capacity would be high [7]. The work of Russel et al. shows that it exists a link between the hBM-MSC multipotence and their capacity to proliferate and form colonies [20]. Indeed, more the hBM-MSC proliferate and form colonies, more the cells can be differentiated into osteocytes, adipocytes and chondrocytes [20]. Others studies showed that neither the age, nor the sex of the donors would affect the differentiation capacity of the hBM-MSC but that their phenotype would be rather involved. In fact, the expression of CD10 and CD119 markers could be correlated in a positive way with the mRNA coding for the transcription factor PPARG (peroxisome proliferator-activated receptor), specific of adipogenic differentiation [11].

However other works showed that the age of the donors could affect the differentiation capacity of hBM-MSC. Indeed, Stolzing et al. showed that osteogenic and chondrogenic differentiation could be decreased in hBM-MSC from aged donors but that adipogenic differentiation would not be modified [8]. Other authors showed that adipogenic and osteogenic differentiation would decrease with the increase in the donors age but no effect was observed in chondrogenic differentiation [27]. Moreover, osteogenic differentiation would decrease with the increase on the donors age [9] and the differentiation capacity of human hBM-MSC in biomaterials made up of hydroxyapatite and implanted in nude mice showed that the osseous formation correlated with the age of the BM-MSC [6]. All the studies of hBM-MSC multipotence coming from donors of different ages were not performed with cells of the same passage, which could explain the contradictory results between the studies.

6. Bone-marrow mesenchymal stromal cells aging, senescence and oxidative stress

Denham Harman was one of the first to have proposed the theory of the oxidative stress oxidizing in aging [28]. In the literature, many works show that cells, often fibroblasts, cultivated in vitro and exposed to oxidative stress or ionizing radiation undergo a stress inducing senescence [29]. However, little works show the effect of the age of the donor on the senescence of the hBM-MSC. Stolzing et al. analyzed the senescence thanks to the inhibitors of the cellular cycle P21 and P53, but without the measure of the β-galactosidase activity, and showed that the expression of these markers was increased besides the level of oxidative stress and the production of NO in the hBM-MSC coming from old donors [8].

Wagner et al. showed that the replicative senescence of hBM-MSC related to their in vitro amplification, is a continuous process starting at the beginning of the culture [14]. This process includes deteriorations in their phenotype, their capacity of differentiation and their profile of gene and microRNA expression. Later, the same team showed that the aging of the hBM-MSC related to the age of the donors involves modifications of gene expression and is associated with their replicative senescence in vitro, indicating thus that hBM-MSC undergo a similar process in vivo [25]. Other studies showed that hBM-MSC coming from patients with lupus are characterized by a senescent phenotype which is controlled by the cell cycle inhibitor protein P16 which activates the protein kinase ERK1/2 (Extracellular signal-regulated kinase). Indeed, in these hBM-MSC, the inhibition of the expression of P16 induces a reduction in the senescence of the cells [30]. The senescent phenotype of the hBM-MSC would also be related to the expression of the inhibitors of the cell cycle P53 and P21 [31]. Moreover, another work shows that the expression of these markers would increase with the years donors of the hBM-MSC [9].

Work of Brandl et al. shows that MSC in vitro exposed to oxidizing stress induced by a sub-lethal amount of H₂O₂ were more resistant (in terms of proliferation and telomeres length) compared to chondrocytes or fibroblasts, but that this tolerance was decreased in the old patients [32]. Other works showed that the hBM-MSC exposed in vitro to ROS (Reactive Oxygen Species) and RNS (Reactive Nitrogen Species) were able to control the oxidative stress thanks to glutathion [33] and that the level of ROS would increase with the increase of the age of hBM-MSC donors.

Moreover, one study of Song et al. showed that the elimination of the ROS could be a new strategy to facilitate the adhesion of the MSC implanted into ischemic myocardium [34].

Lastly, works performed on hBM-MSC show that the endogenous level of ROS was correlated with the level of cell senescence markers and that the surexpression of APE1/Ref-1 (apurinic/apyrimidinic endonuclease1/redox Factor-1) decreased the senescent phenotype of hBM-MSC [35]. APE1/Ref-1 is known to inhibit the production of ROS by inhibiting GTPase Rac1 itself controlled by the NADPH oxidase. This work shows that the maintenance of the characteristics of the hBM-MSC while controlling the production of ROS and the senescence are parameters important to take into account for the use of these cells in regenerative medicine.

7. Bone-marrow mesenchymal stromal cells aging and telomere length

Many studies suggest that telomeres shortening, TTAGGG double-stranded sequences repeated and present at the ends of chromosomes are associated with the many pathologies including those related to age [29,36]. The telomerase activity, the enzyme responsible for the synthesis of telomeres was described in cancer cells, in the cells of the germinal lines, and in embryonic stem cells (ESC) [19]. However, it was shown that the activity of this enzyme was reduced during differentiation of mouse ESC [37]. The enzyme is also active in adult stem cells such as the hematopoietic stem cells (HSC) [19] and the telomere shortening process was proposed to limit the renewal of the HSC [36]. Concerning the activity of the telomerase on adult stem cells such as hBM-MSC, that remains under discussion in the literature [19]. Indeed certain studies showed the absence of telomerase activity in hBM-MSC [38,39] while work of our team shows that the hBM-MSC have a very weak telomerase activity compared to control cancer cells “Hela” [40]. Other work of Zimmermann et al. shows that this weak telomerase activity would be found in very rare sub-population of hBM-MSC [41].

Moreover the work of Baxter and colleagues [3], shows an in vitro reduction in the telomeres length dependent on the age of the donors of hBM-MSC. They also show that it exists a correlation between the capacities of proliferation of the hBM-MSC and the telomere length of the cells in culture and the age of the donors. Indeed an equivalent telomeric erosion is observed in vitro with each cell division whatever the age of the donors. However, telomere length is significantly higher for hBM-MSC from young donors (0–18 years) compared to hBM-MSC from old donors after 16 PD population doublings. This difference is not significant any more when the growth is stopped when cells reached senescence. These results suggest that telomere erosion could occur in vivo.

Lastly, work of Simonsen et al. shows that the overexpression of a non-endogenous telomerase in human hBM-MSC removes the senescent phenotype and maintains the cellular functions such as the proliferation and the multipotence, including osteogenic differentiation [38]. A better comprehension of telomere erosion and telomerase activity in hBM-MSC would make it possible to improve the therapeutic strategies with MSC.

8. Conclusion

In the last years the knowledge of BM-MSC properties increased and different clinical trials were developed to test the performances of these cells in a large spectra of diseases. A large number of that diseases are age-related ones, doing the elderly people a target population for this kind of therapeutic approaches. Even if not all the researches showed an effect of MSC aging on cell properties, most of them postulated that this effect there exists. Consequently it is very important to study MSC in the context of aging to find parameters able to characterize aged MSC, which could be nonfunctional cells impacting therapeutic effectiveness.

Conflict of interest

The authors have no conflict of interest to report.