Influence of age on stem cells depends on the sex of the bone marrow donor

Abstract Ageing is often accompanied by an increase in bone marrow fat together with reduced bone volume and diseases of the bone such as osteoporosis. As mesenchymal stem cells (MSCs) are capable of forming bone, cartilage and fat tissue, studying these cells is of great importance to understand the underlying mechanisms behind age‐related bone diseases. However, inter‐donor variation has been found when handling MSCs. Therefore, the aim of this study was to investigate the effects of donor age and sex by comparing in vitro characteristics of human bone marrow‐derived MSCs (hBMSCs) from a large donor cohort (n = 175). For this, hBMSCs were analysed for CFU‐F capacity, proliferation, differentiation capacity and surface antigen expression under standardized culture conditions. The results demonstrated a significantly reduced CFU‐F number for hBMSCs of female compared to male donors. Furthermore, there was a significant decrease in the proliferation rate, adipogenic differentiation potential and cell surface expression of SSEA‐4, CD146 and CD274 of hBMSCs with an increase in donor age. Interestingly, all these findings were exclusive to hBMSCs from female donors. Further research should focus on postmenopausal‐related effects on hBMSCs, as the results imply a functional loss and immunophenotypic change of hBMSCs particularly in aged women.


| INTRODUC TI ON
One of the current major challenges in trauma surgery is the treatment of larger bone defects and cartilage damage caused by arthrosis or traumatic incident.
In addition, various studies have already shown that progressing age correlates adversely with the repair and regeneration potential of the human body. 1,2 Furthermore, there is an increasing prevalence of musculoskeletal diseases in our ageing society. 3 As a result, interest in the therapeutic potential of human bone marrow-derived mesenchymal stem cells (hBMSCs) has increased enormously in recent years. In this context, tissue replacement of bone and cartilage by using hBMSCs is a promising therapy option in regenerative medicine. However, while successful application of autologous MSCs has | 1595 SELLE Et aL. been reported in some patients, 4,5 other studies could not find clear evidence for effective treatment with hBMCs in surgery. 6,7 These contradictory results may be explained by the functional disparity observed in MSCs from different donors, including their proliferation and differentiation capacity. 8 Some authors documented an inverse relationship between age and CFU-F capacity 9-14 as well as proliferation rate, 11,[15][16][17] while others found no significant evidence for such a relationship. [18][19][20][21] Furthermore, the expression level of some cell surface antigens, such as CD146, 22,23 CD274, 23 SSEA-4 24 and others, 23 was reported to be associated with donor age.
Similarly, a correlation between the regenerative potential and the sex of MSC donors has been investigated. Some authors have found a negative correlation for female donors regarding osteogenesis and the production of collagen type I. [25][26][27][28] Furthermore, a correlation between oestrogen and osteogenesis with an inverse correlation to adipogenesis has been described. 29 Other authors could not find any differences for single-cell cloning efficiency, cumulative population doubling or colony-forming unit-fibroblast (CFU-F) Assays regarding different sex. 15,21,30 One reason for these contradictory results could be the heterogeneous study design of the studies conducted so far. Most of the relevant studies used BMSCs originating from either humans, mice, rats or dogs for the experiments, but some also worked with adipose tissue-derived MSCs. Furthermore, different isolation and purification techniques with different scoring criteria were used. A decisive step towards a comparable study situation was taken by the Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy (ISCT). 31 Here, standardized minimal criteria were introduced to characterize and define mesenchymal stem cells: MSCs have to be plastic adherent when maintained under standard culture conditions. MSCs must express CD105, CD73 and CD90 with no expression of CD45, CD34, CD14 or CD11b, CD79a or CD19 and HLA-DR surface molecules. In addition, MSCs must differentiate into osteoblasts, adipocytes and chondrocytes in vitro. 31 Therefore, in this study, we aimed to investigate the effects of donor age and sex on the in vitro characteristics of hBMSCs utilizing a large donor cohort (n = 175). For this, hBMSCs from donors

| Colony-forming unit-fibroblast assays and growth rate
Two CFU-F assays were performed to investigate the self-renewal potential of the cells. The first CFU-F assay was performed with hBMSCs in P1, the second with hBMSCs in P3. For this, the hBM-SCs were seeded at three different densities (125, 250 and 500 per well) as duplicates in 6-well plates (Greiner Bio-one, Frickenhausen, Germany). After an incubation period of 10 days at 37°C and 5%

| Differentiation
Adipogenic, chondrogenic and osteogenic differentiation assays were performed in P4. Therefore, cryo-conserved hBMSCs were thawed and expanded in complete medium. Then, 150,000 cells per 9.6 cm² were seeded into 6-well plates. After 24 h, the complete medium was replaced with the according differentiation or con- including 20 mM HEPES, 1% P/S, 0.1 μM dexamethasone, 10 μL/mL insulin/transferrin/selenium (Sigma-Aldrich), 170 µM ascorbate-2phosphate, 1 mM sodium pyruvate (Biochrom), 350 μM proline (Carl Roth, Karlsruhe, Germany) and 10 ng/mL transforming growth factor beta-3 (TGF-β3; PeproTech). The control medium did not contain TGF-β3. The media for all the groups were replaced every 7 days, and the differentiations were stopped after 28 days. to the total area of the pellets. All analyses were performed blindly without prior knowledge to underlying donor data.
More detailed information about each antibody is listed in Table S3.

| Statistics
The statistical analyses were performed with IBM SPSS Statistics software (version 26). The data were first tested for distribution using the Shapiro-Wilk test and found to be non-normally distrib-   Figure 2G). There was a significant positive correlation between doubling time and donor age in hBMSCs from female donors (r = 0.388, p < 0.001; Figure 2H), but not in hBMSCs from male donors (r = 0.139, p = 0.187; Figure 2I).

| Differentiation potential
The hBMSCs were able to differentiate in vitro according to three lineages ( Figure 3A, 4A, 5A) to varying degrees.
The capacity of the cells to differentiate into adipocytes in vitro was comparable between hBMSCs from female and male donors  Figure 5B). Furthermore, there was no significant correlation of osteogenesis and age, neither in hBMSCs from female ( Figure 5C) nor from male donors ( Figure 5D).

| Surface antigen markers
The hBMSCs were screened for the surface expression of 34 distinct antigens in P4 via flow cytometry in search for sex-and ageassociated differences in their surface expression profiles (Tables 1,   S4). CD13, CD29, CD44, CD73, CD90, CD105, CD166 and GD2 were highly expressed (Mdn > 90%) irrespective of age or sex. There Correlation analyses revealed that some of these surface antigens were found to be expressed in an age-dependent manner,  Figure 6G). Yet, on hBMSCs from female donors the expression of CD274 negatively correlated with donor age (r = −0.273, p = 0.021; Figure 6H), while once again on hBMSCs from male donors no significant correlation was found (r = −0.127, p = 0.234; Figure 6I).
Beyond that, further statistical significances were uncovered, although with differences in expression levels smaller than 5%, which are not listed in this manuscript as they are of no clinical relevance

| DISCUSS ION
The aim of this study was to assess inter-donor variability of hBM-SCs based on donor age and sex.
The decline in CFU-F potential of hBMSCs over passages in culture as observed in the present study by comparing CFU-F forming capacity between P1 and P3 is in agreement with studies by other authors and has been associated with accelerated culture-induced in vitro ageing of MSCs, for example by shortening of telomeres and a higher proportion of senescent cells with continued population doublings, as reviewed by Ganguly et al. 1 The present study shows no correlation between donor age and CFU-F capacity, fitting the results obtained by few others. 18 34 This, in turn, might signify that donor age and sex are of less significance when using hBMSCs as therapeutic agents in regenerative strategies.
In our study, SSEA-4, CD146 and CD274 expression on hBM-SCs decreased with an increase in donor age. This is mainly in line with findings by other authors, as SSEA-4, CD146 and CD274 have already been acknowledged as markers of MSC ageing in the literature. [22][23][24]45 Interestingly, the present study is the first to suggest that this effect may be more pronounced in hBMSCs from female donors. SSEA-4 was first proposed as a potential marker for a po-

ACK N OWLED G EM ENTS
We would like to thank all bone marrow donors without whom this study would not be feasible. Furthermore, we would like to thank

CO N FLI C T S O F I NTE R E S T S
The authors declare no conflicts of interests. Writing -review & editing (lead).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.