Sublethal Oxidative Stress Induces the Premature Senescence of Human Mesenchymal Stem Cells Derived from Endometrium

The specific responses of mesenchymal stem cells to oxidative stress may play a crucial role in regulation of tissue homeostasis as well as regeneration of organs after oxidative injury. The responses of human endometrium-derived mesenchymal stem cells (hMESCs) to oxidative stress remain still unknown. Herein, we examined the impact of H2O2 on cell viability, induction of premature senescence, and apoptosis. hMESCs were highly resistant to H2O2 compared with human diploid fibroblasts. To test a hypothesis whether hMESCs may undergo oxidative stress-induced premature senescence, cells were briefly exposed to the sublethal H2O2 doses. H2O2-treated cells were permanently arrested, lost Ki67 proliferation marker, and exhibited a senescent phenotype including cell hypertrophy and increased SA-β-Gal activity. Additionally, in stressed cells the expression levels of p21Cip1, SOD1, SOD2, and GPX1 were elevated. hMESCs survived under stress were not able to resume proliferation, indicating the irreversible loss of proliferative potential. While the low H2O2 doses promoted senescence in hMESCs, the higher H2O2 doses induced also apoptosis in a part of the cell population. Of note, senescent hMESCs exhibited high resistance to apoptosis. Thus, we have demonstrated for the first time that hMESCs may enter a state of premature senescence in response to sublethal oxidative stress.


Introduction
Stress responses of human embryonic and adult stem cells to -radiation, oxidative stress, heat shock, and so forth are widely researched to establish cell-based strategies of tissue repair, tissue engineering, and transplantation [1]. Human mesenchymal stem cells are adult multipotent stem cells with the capacity of self-renewal and undergoing adipogenic, osteogenic, chondrogenic, myogenic differentiation [2,3]. They contribute to the homeostatic maintenance of many organs and tissues [4,5]. Unlike some other adult stem cells (e.g., hematopoietic stem cells) human mesenchymal stem cells are not immortal. These cells exhibit ex vivo growth characteristics typical of the Hayflick model of cellular senescence with a limited life span [6]. Recently, it has been reported that the mesenchymal stem cells subjected to oxidative stress [7][8][9] or ionizing radiation [10][11][12] may undergo stress-induced premature senescence in vitro. Many types of normal and tumor cells also enter a state of premature senescence after exposure to radiation [13][14][15], H 2 O 2 [16][17][18][19], or treatment with histone deacetylase inhibitors [20,21]. Prematurely senescent cells exhibit some of the characteristics inherent in replicatively senescent cells, including a large flat morphology, increased senescence-associated -galactosidase (SA--Gal) activity, and permanent cell cycle arrest [14,17]. Besides, cellular overactivation and hyperfunction, feedback signal resistance, and loss of regenerative potential are considered hallmarks of senescence [22]. Progress in understanding the causes and mechanisms of cellular senescence and significance of senescence for ageing and suppressing cancer has been reviewed [23][24][25].
In the current study, oxidative stress responses of human mesenchymal stem cells derived from endometrium (hMESCs) were investigated. Our knowledge of specific responses of these cells to stress is very limited, though they prove to be useful in the treatment of pathologies in which tissue damage is linked to oxidative stress. Unlike most of the human mesenchymal stem cells, the isolation of which as a rule is complicated by invasive procedures, the mesenchymal stem cells produced from desquamated endometrium in menstrual blood by a simple noninvasive way provide a good opportunity to explore the stress responses of hMESCs. Regarding hMESCs, phenomenon of premature senescence induced by oxidative stress remains still unknown. This study aimed to test a hypothesis whether hMESCs after exposure with sublethal doses of H 2 O 2 may undergo the stress-induced premature senescence. In parallel, the impact of H 2 O 2 on cell viability and development of apoptosis has been evaluated.

Materials and Methods
2.1. Cell Culture and Cell Treatment. Human mesenchymal stem cells isolated from desquamated endometrium in menstrual blood (hMESCs, line 2304), as described previously [26], as well as human embryonic lung-derived diploid fibroblasts (HDF, line FRL-9505) were cultured in complete medium (DMEM/F12 (Gibco BRL, Gaithersburg, MD, USA) supplemented with 10% FBS (HyClone, Waltham, MA, USA), 1% gentamycin, and 1% glutamax (Gibco BRL, Gaithersburg, MD, USA)) at 37 ∘ C in humidified incubator, containing 5% CO 2 . hMESCs have a positive expression of CD73, CD90, CD105, CD13, CD29, and CD44 markers and absence of expression of the hematopoietic cell surface antigens CD19, CD34, CD45, CD117, CD130, and HLA-DR (class II). Multipotency of isolated hMESCs is confirmed by their ability to differentiate into other mesodermal cell types, such as osteocytes and adipocytes. Besides, the isolated hMESCs partially (over 50%) express the pluripotency marker SSEA-4 but do not express Oct-4. Immunofluorescent analysis of the derived cells revealed the expression of the neural precursor markers nestin and beta-III-tubulin. This suggests a neural predisposition of the established hMESCs. These cells are characterized by high rate of cell proliferation (doubling time 22-23 h) and high cloning efficiency (about 60%). Cells at early passages (between 6 and 9 passages for hMESCs and between 16

Assessment of Cell
Viability. The cell viability after exposure to H 2 O 2 for 1 h was evaluated by the enzymatic conversion of MTT (AppliChem, Darmstadt, Germany, number A2231) to formazan in live cells. The culture medium from the cells grown in plates was removed, and 3-(4.5-dimethylthiazol-2-yl)-2.5-diphenyltetrazolium bromide (MTT; 0,715 mg/mL) in serum-containing growth medium was added to each well. In 2 h the solution was changed to DMSO to solve formazan produced. The plates were shaken for 15 min at room temperature; thereafter the absorbance was measured at 570 nm using microplate reader (Fluorofot "Charity, " Russia). All points were read as parallels of 8 similar samples. The average absorbance at a given time point was normalized to the start time point.

Flow
Cytometry. Adherent cells were rinsed twice with PBS and harvested by trypsinization. Detached cells were collected with supernatants, pelleted by centrifugation. Detached and adherent cells were finally pooled and resuspended in PBS. One part of each sample was used for propidium iodide (PI) staining to evaluate cell viability and another part for cell cycle phase distribution analysis that was performed as described previously [27]. 50 g/mL PI was added to each sample just before analysis and mixed gently. Samples were analyzed on a Coulter EPICS XL Flow Cytometer (Backman Coulter, Brea, CA, USA). For cell cycle analysis, each cell sample was suspended in 300 L PBS containing 200 g/mL of saponin (Fluka, NY, USA), used for cell permeabilization, 250 g/mL RNase A (Sigma, St. Louis, MO, USA, number R4642), and 50 g/mL PI, incubated for 30 min at room temperature and subjected to FACS analysis. At least 10,000 cells were measured per sample. Cell cycle analysis was performed using Win MDI program version 2.8 and ModFit LT software (Verity Software House, Topsham, ME, USA).

FACS Analysis of Cell Enlargement.
The same procedure of sample preparation as described previously was done for light-scattering cytometry. As after H 2 O 2 treatment live and dead cells were very close to each other on one-parameter histogram, two-parameter histogram was used (FL4LOG versus FSLOG) to discriminate live and dead cells. Analysis of each sample was performed for 100 sec with high sample delivery. Cell size of control and H 2 O 2 -treated cells was measured by means of cytometric light scattering of PIstained cells by using Win MDI program version 2.8.

SA--Gal Activity. Cells expressing senescent-associated
-galactosidase were detected with senescence -galactosidase staining kit (Cell Signaling Technology, Beverly, MA, USA, number 9860) according to manufacturer's instructions. The kit detects -galactosidase activity at pH 6 in cultured cells which is present only in senescent cells and is not found in presenescent, quiescent, or immortal cells. The percent of SA--Gal-positive cells was calculated by counting not less than 500 cells.
2.6. Immunofluorescence Staining. Cells cultured on coverslips were fixed with PBS/4% formalin for 15 min and then permeabilized with 0.1% Triton X-100. After blocking with 1% bovine serum albumin, they were incubated with a rabbit polyclonal antibody against Ki67 (Abcam, Cambridge, UK, number 15580) (1 : 1000) overnight at 4 ∘ C and then with Alexa Fluor 568 donkey anti-rabbit antibody (Invitrogen, Carlsbad, Oxidative Medicine and Cellular Longevity 3 USA, number A10042) (1 : 500) at room temperature for 1 h after extensive washing with PBS/0.1% Tween 20 between each step. The slides were counterstained with 1 g/mL DAPI (Sigma, St. Louis, MO, USA, number D9564) and mounted using 2% propyl gallate. A Zeiss Axiovert 200M fluorescence microscope (Carl Zeiss, Germany) equipped with a digital camera DFC 420C (Leica, Germany) utilizing Adobe Photoshop software was used to view and acquire images.

Apoptosis
Detection. Apoptosis detection was performed by Annexin V/PI staining according to standard manufacture protocols (BD Pharmingen). H 2 O 2 -treated and control cells were harvested as described previously. Then cells were washed twice with cold PBS, resuspended in 1X binding buffer at concentration 10 6 cells/mL. 100 L of this suspension was transferred to 5 mL tube, and 5 L Annexin V/FITC (Annexin V/FITC Apoptosis Detection Kit II, BD Biosciences, San Diego, CA, USA) and 10 L PI were added; suspension was gently mixed and incubated for 15 min at room temperature in the dark. Just before flow cytometric analysis 400 L of 1X binding buffer was added to each sample.
2.10. Statistics. All data are presented as the mean and standard error of the mean from at least three separate experiments performed. Statistical differences were calculated using the Student's -test and considered significant at * < 0.05.  [28,29]. On the other hand, it has been recently reported that some types of human mesenchymal stem cells are very sensitive to H 2 O 2 exposure [9]. Susceptibility of hMESCs to oxidative stress remains unexplored up to date. We earlier reported that hMESCs subjected to prolonged treatment (for 24 h) with H 2 O 2 demonstrated a higher resistance compared with human diploid fibroblasts [30]. In this study, a pulse cell treatment with H 2 O 2 in varying concentrations from 200 M to 2 mM for 1 h was applied. Human diploid fibroblasts (HDF) were used as a H 2 O 2 sensitive cell model to compare effect of H 2 O 2 cytotoxicity on hMESCs. Firstly, it was necessary to assess hMESCs viability under oxidative stress to examine a sublethal H 2 O 2 concentration required for further experiments. Previously we have found out that, at a fixed H 2 O 2 concentration, both cytotoxicity and rate of degradation were dependent on the volume of H 2 O 2 solution added to the culture medium; therefore the volume used was adjusted proportionally according to the surface area in order to obtain a consistent H 2 O 2 cytotoxicity. In addition, it was very important to control plating cell density because the H 2 O 2 effect (at equal volume and concentration) on cells was inversely related to cell density; that is, confluent cell cultures were more resistant to H 2 O 2 than subconfluent ones. Cell viability was evaluated by MTT assay as a broad indicator of cellular activity that allows estimating a number of viable cells via monitoring mitochondrial dehydrogenase activity. As shown in Figure 1 Figure 2). Consequently, the significant insensitivity to H 2 O 2 was consistent with the enhanced expression levels of the antioxidant enzymes. These findings are in agreement with previous report demonstrating a high resistance of human bone marrow-derived mesenchymal stem cells to oxidative stress [29]. The conflicting findings demonstrating a particular sensitivity of human umbilical cord blood-derived mesenchymal stem cells to H 2 O 2 have correlated with the low levels of antioxidant enzyme activity [9].        Figure 4(a), the pattern of growth curves indicates a significant increase (more than two times in 5 days) in the number of proliferating control cells compared with H 2 O 2 -treated cells. Consequently, 200 M H 2 O 2 caused a permanent growth arrest, that is, a permanent loss of the proliferative potential. Additionally, a proliferative status of cells was examined by staining with antibodies against proliferation marker Ki67. As seen in Figure 4(e), in 5 days after H 2 O 2 treatment, there were no Ki67-positive cells in the cell culture, while the proliferating control cells had a pronounced staining. As viability of hMESCs in response to 200 M H 2 O 2 did not decrease appreciably (Figure 1), we suggested that H 2 O 2 -induced growth inhibition of hMESCs could be associated with rather the cell cycle arrest than promotion of cell death. The analysis of the cell cycle phase distribution in hMESCs showed that a pulse H 2 O 2 treatment led to the arrest in all of the cycle phases (Figure 4(b), upper panel). Treated cells demonstrated the prolonged arrest, at least, for 5 days. The phase distribution of treated cells in each time point tested was characterized with a minor accumulation of cells in G0/G1 phase compared with control cells. The distribution analysis with using light scattering confirmed these findings (Figure 4(b), lower panel). To test whether arrested cells could recover their proliferative potentials, in 2 days after treatment cells were reseeded and cultivated for 3 more days. As expected, reseeded cells also displayed the cell cycle arrest. Moreover, cell cycle phase distributions of both reseeded    and taken-before-reseeding cells were identical (Figure 4(b)). Consequently, the senescent cells were not able to resume proliferation even after being reseeded, indicating the irreversible growth arrest. These observations were confirmed by proliferation assay (Figure 4(a)), which demonstrated that H 2 O 2 -treated cells were not able to proliferate normally for 5 days. Overall, these results suggest that cellular senescence in hMESCs was induced through growth arrest by H 2 O 2 .

A Permanent Loss of the Proliferative Potential in hMESCs
Is Accompanied with Elevated Levels of p21. In human mesenchymal stem cells, cyclin-dependent kinase inhibitor p21 was recently shown to be upregulated during H 2 O 2 -induced premature senescence [7,8]. Increased levels of p21 may mediate the initiation of H 2 O 2 -induced cell cycle arrest by inhibiting various cyclin-dependent kinases that contribute cell cycle phase progression [14,16]. To find out whether p21 could be involved in the regulation of H 2 O 2 -induced senescence of hMESCs, protein and mRNA expression levels of p21 were determined. 200 M H 2 O 2 promoted a significant elevation in protein (Figure 4(c)) and mRNA (Figure 4(d)) expression of p21 in 7 h after treatment. An inducible expression of p21 was upregulated during 1-2 days with a following decline to insignificant, but not control, levels and was accompanied with the cell cycle arrest at the same time (Figure 4(b)). Importantly, the arrested cells thereafter could acquire a senescent morphology (Figure 3(a)) but could not resume proliferation (Figure 4(a)). We assume that the elevated p21 expression is essential to drive H 2 O 2induced premature senescence in hMESCs. In support of our findings, it has been reported that, in bone marrow-derived mesenchymal stem cells exposed to sublethal doses of H 2 O 2 , a rapid decrease of proliferation rate was detected within 3 days and correlated with G1 phase arrest of the cell cycle when p21 was accumulated at the same time [8].
In summary, our findings strongly indicate that hMESCs under a sublethal oxidative stress are able to undergo premature senescence.  preliminary results, apoptosis was mediated by both caspase-8 and activated caspase-3; however, the exact mechanism of H 2 O 2 -induced apoptosis in hMESCs remains to be further elucidated.

Effect of
Interestingly, 900 M H 2 O 2 triggered not only delayed apoptosis but also led to the emergence of enlarged and flattened cells in the same cell cultures. Cell changes in the presence of both 200 M and 900 M H 2 O 2 were similar in appearance. These observations prompted us to test if cell hypertrophy could be connected with premature senescence. As shown in Figure 6, 900 M H 2 O 2 actually promoted the senescent morphology and SA--Gal staining, permanent growth arrest, and approximately 2fold increase of cell size, pointing to premature senescence of the main part of cell population. Notably, the major senescence features induced by both 200 M and 900 M H 2 O 2 were found to be alike. Together, these findings demonstrate that hMESCs exhibit high apoptosis resistance compared with human mesenchymal stem cells derived from both umbilical cord blood [9] and bone marrow [7], in which H 2 O 2 above 200 M triggered apoptosis, whereas 100-150 M H 2 O 2 induced senescence. Moreover, after exposure to sublethal doses of H 2 O 2 , senescent hMESCs acquired the increased stability in culture and displayed enhanced resistance to H 2 O 2 -induced apoptosis (data not shown). Many cell types acquire resistance to some apoptotic signals when they become senescent. So, senescent human fibroblasts resist apoptosis induced by oxidative stress or growth factor deprivation but do not resist Fas-mediated apoptosis [31,32]. Resistance to apoptosis might in part explain the enhanced stability of senescent cells in culture. The mechanisms by which senescent cells resist apoptosis are poorly investigated. The senescence and apoptosis regulatory systems are supposed to communicate probably through their common regulator, p53 tumor suppressor protein [33].
In this study, we have provided the reliable evidence for our hypothesis that hMESCs are able to undergo the premature senescence in response to oxidative stress induced by H 2 O 2 in a wide range of concentrations from 200 to 900 M. According to data obtained, entering senescence was accompanied with a rapid initiation of the cellular events, such as the changes of cell phenotype, the increase of SOD1, SOD2, and GPX1 expression, and upregulation of p21 without increase over time, leading to the irreversible cell cycle arrest and loss of proliferative potential. Since 2009, when phenomenon of stress-induced premature senescence in human mesenchymal stem cells was described for the first time, there were only a few publications concerning the oxidative stress-induced premature senescence of human mesenchymal cells derived from bone marrow [7,8] and umbilical cord blood [9]. Even though both stem cell lines under sublethal stress respond with senescence, the major features of this process, in particular, dynamics of p21 accumulation and decline of cell proliferation rate, were extremely different, depending on the cell context. The precise molecular mechanism required to regulate the oxidative stressinduced premature senescence of human mesenchymal stem cells is far from understanding. Senescence program seems to develop in mesenchymal stem cells as a result of DNA damage response, leading to functional activation of either the p53/p21 or the p16INK4a (p16)/retinoblastoma protein (pRb) pathway, both of which can establish and maintain the growth arrest that is typical of senescence [23,24]. The cyclin-dependent kinase inhibitors p16 and p21 may maintain pRb in active hypophosphorylated state [34,35]. In turn, pRb halts cell proliferation by suppressing the activity of transcription factor E2F that regulates cell cycle progression. Our preliminary data, indicating a time-and dose-dependent formation of foci that contain phosphorylated histone H2AX ( H2AX), activation of both ATM and p53, and upregulation of p21 expression, suggest that in hMESCs subjected to sublethal doses of H 2 O 2 the senescence process may be controlled by the p53 pathway. In parallel, by monitoring the kinetics of p38 mitogen-activated protein kinase (MAPK) activation in H 2 O 2 -induced senescence of hMESCs, we have revealed a rapid and continued phosphorylation of p38 MAPK, indicating its possible role in the regulation of the premature senescence [36]. On the other hand, pRb was reported to induce growth arrest as a downstream molecule of p38 MAPK [37]. Taking into consideration these results, we cannot exclude the possibility that p16/pRb signaling cascade is also implicated in hMESCs senescence promotion.
Understanding the mechanisms of senescence process will be of great importance in developing applications of hMESCs in regenerative medicine to provide new strategies in autologous transplant and bioengineering. Primarily, hMESCs may be applied for cell therapy of infertility associated with decidualization insufficiency. Decidualization of endometrium is known to be an essential process for embryo implantation, placenta forming, and maintenance of pregnancy [38]. A noninvasive and easily available source for isolation of hMESCs, high proliferation activity during longterm cultivation, genetic stability, lack of tumorigenicity [39], and low immunogenicity make hMESCs a promising source of stem cells for clinical applications, including reproduction technology.
In summary, we have displayed for the first time that hMESCs in oxidative stress conditions undergo a premature senescence. Data obtained broaden a conception of mesenchymal stem cell senescence under oxidative stress. Taken together, the findings presented here and the data published allow us to assume that induction of premature senescence might be a common physiological response to sublethal oxidative stress in human mesenchymal stem cells of any origin.