Protocols for the study of stem cells from with patients’ specimens were approved by the Ethics Committee of Biomedical Research of Vilnius District, No. 158200-18/7-1049-550. EndSCs and MenSCs were obtained from females experiencing unexplained infertility. 12 donors were used for the experiments with endometrial stem cells and six donors for experimentations with menstrual blood stem cells. Endometrial stem cells were isolated as described previously (Tavakol et al., 2018) from scratches of the inner wall of endometrial lining and menstrual blood stem cells were obtained by collecting menstrual blood using menstrual cup. MenSCs were isolated as described by Skliute et al. (2021). Endometrium and menstrual blood mesenchymal stem cells were seeded into 25 cm² plastic cell culture flasks at 1.6 × 10 ⁴ cells/cm ² density and maintained in 37°C, 5% СO₂ incubator. The cells were cultivated in the growth medium DMEM/F12 (DMEM, Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% Fetal Bovine Serum (FBS) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), 100 U/ml penicillin and 100 μg/ml streptomycin (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). The medium was replaced every 2 days, and when the confluence reached 90–100%, the cells were detached from the surface using trypsin-EDTA (0.05%) solution (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) and reseeded into new cell culture flasks. EndSCs and MenSCs undergo around 2–12 divisions for early passage cells (2p.–7p.) and around 28–40 for late passage cells (15p.–19p.). In gene expression and Western blot protein analysis AF-MSCs were used as a control group. AF-MSCs were isolated from five donors and cultivated according to (Savickiene et al., 2015) protocol. Cells were seeded at 1.6 × 10⁴ cells/cm² density and passaged at approximately 80–90% confluence. AF-MSCs at passage three were then selected for following experiments. During cultivation of EndSCs and MenSCs we assessed proliferation and viability every time the cells were passaged. This was done by trypan blue exclusion test. The cells were trypsinized, centrifuged and resuspended in DMEM/F12 medium. Then equal volumes of cells and trypan blue 0.4% (Pharmacia LKB, Uppsala, Sweden) were mixed together and analyzed in a hemocytometer under a light microscope. The cells were counted according to their color—viable cells were transparent and dead cells were colored blue, and depending on the number of cells, viability was evaluated. EndSCs and MenSCs doubling time (DT) was assessed by monitoring cell proliferation during passaging and was calculated using formula: DT = duration ( days ) × log ( 2 ) log ( final concentration ) − log ( initial concentration )

For phenotypical characterization of undifferentiated EndSCs and MenSCs, cell surface markers were analyzed. For one assay around 5 × 10⁴–6 × 10⁴ cells were collected by trypsinization and centrifugation at 400 ⨯ g for 5 min. Supernatant was discarded and cells were suspended in 50 μl phosphate-buffered saline (PBS) solution (Genaxxon bioscience, Ulm, Germany) and incubated with antibodies against cell surface markers for 30 min on ice in the dark. After incubation samples were washed twice with PBS. The cells were then resuspended in 100 μl PBS and fixated with 100 μl 2% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) as final volume. Typical mesenchymal stem cells markers were detected using APC conjugated mouse anti-human antibodies against CD44 (Biolegend, CA, USA), CD73 (Exbio, Vestec, Czech Republic), CD90 (Exbio, Vestec, Czech Republic) and CD105 (Exbio, Vestec, Czech Republic). Markers for hematopoietic progenitor cells and endothelial cells were also analyzed using APC conjugated mouse anti-human antibodies against CD34 (Biolegend, CA, USA) and FITC conjugated mouse anti-human antibodies against CD45 (BD Pharmingen, CA, USA). Next PE-conjugated mouse anti-human antibodies against Activated Leukocyte Cell Adhesion Molecule (ALCAM) CD166 (Biolegend, CA, USA), APC conjugated mouse anti-human antibodies against c-kit stem cell factor receptor CD117 (Exbio, Vestec, Czech Republic) and against melanoma cell adhesion molecule CD146 (Biolegend, CA, USA). For isotype controls mouse IgG1-APC (Biolegend, CA, USA, AND, Exbio, Vestec, Czech Republic), IgG1-PE (Biolegend, CA, USA), IgG2a-APC (Exbio, Vestec, Czech Republic) and IgG1-FITC (Exbio, Vestec, Czech Republic) were used. Labeled samples were then measured using Partec flow cytometer (Sysmex Corporation, Cobe, Hioko, Japan) with Flowing Software 2 software.

Early (2p.–7p.) and late (15p.–19p.) passages of endometrial and menstrual blood mesenchymal stem cells were selected to undergo decidualization in vitro. Decidualization was induced by culturing cells with 0.5 μM dibutyryl cyclic-AMP (db-cAMP) (BioGems International, Inc., Westlake Village, CA, USA) and 1 μM medroxyprogesterone acetate (MPA) (Cayman Chemical Company, Ann Arbor, MI, USA) agents at 1.5 × 10⁴ cells/cm² density. Decidualization medium, used for whole differentiation period consisted of: RPMI (phenol free) medium (Corning Incorporated, NY, USA), 2% Charcoal Stripped Fetal Bovine Serum (Gibco, Thermo Scientific, NY, USA) and 50 μg/ml primocin (Sigma Aldrich, MO, USA). Selected decidualization time was 3 and 6 days, cells were cultured under selected conditions of 37°C, 5% CO₂ in an incubator. During decidualization every other day the medium was replaced with the same fresh medium and filtered through 0.2 μm filter. Undifferentiated EndSCs and MenSCs were used as a control group in later experiments.

RNA from differentiated and undifferentiated EndSCs and MenSCs was isolated by using TRIzol^® reagent (Zymo Research, CA, USA), chloroform and isopropanol for RNA precipitation. After this cDNA was synthesized with SensiFAST cDNA Synthesis Kit (Bioline, London, UK) by following manufacturers recommendations. RT-qPCR was performed using SensiFAST SYBR NoRox Kit (Bioline, London, UK) and Rotor-Gene 6000 Real-time Analyzer (Corbett Life Science, QIAGEN, Hilden, Germany) with experimental conditions as follows: 95°C 2 min, 95°C 5 s, X °C (primer melting temperature) 10 s, 72°C 20 s. RT-qPCR was repeated for 40 cycles for all genes and samples. Used primers are listed in the Supplementary Table S1. The acquired data from RT-qPCR was further analyzed using ∆∆Ct method. By using this method every sample was compared to undifferentiated control and normalized using GAPDH gene expression and therefore relative gene expression was calculated.

Proteins from endometrium and menstrual blood stem cells early and late passages were extracted first by using 0.05% trypsin-EDTA (Gibco, Life Technologies, NY, USA) solution on cells. Collected cells (1–1.5 × 10⁶ cells for one sample) are then washed 2 times with ice-cold PBS solution (Genaxxon bioscience, Ulm, Germany) and incubated with benzonase (Merck, Darmstadt, Germany) for 30 min on ice. After this, 1 volume of 2X SDS lysis buffer (62.5 mM Tris, pH 6.8, 100 mM DTT, 2% SDS, 10% glycerol, and traces of bromphenol blue) was added and finally 10 volumes of 1X SDS lysis buffer was mixed into all samples. Then samples were homogenized with a syringe (26 G), incubated for 5 min at 96°C and centrifuged at 20,000 ⨯ g for 15 min, at 4°C. The supernatant was collected and used for SDS-PAGE electrophoresis or stored at −20°C. The proteins were fractionated in 7.5–15% gradient polyacrylamide gels, transferred onto PVDF membrane (Immobilon P; Millipore, Billerica, MA, USA). Western blot analysis was performed using antibodies against cyclin E1 (Merck Millipore, Burlington, MA, JAV), H3K27me3 (Merck Millipore, Burlington, MA, JAV), EZH2 (Thermo Fisher Scientific, Waltham, MA, USA), SUZ12 (Cell Signaling Technology, MA, USA) and p53 (Santa Cruz Biotechnology, Texas, USA). Antibodies against GAPDH (Abcam, Cambridge, UK) were used as a protein loading control. To achieve higher sensitivity secondary horseradish peroxidase-conjugated antibodies against mouse or rabbit antibodies (DAKO, Glostrup, Denmark) were used. Target proteins were enhanced using chemiluminescence detection with Clarity Western ECL Substrate (Bio-Rad Laboratories, CA, USA). Detection was performed using ChemiDocTM XRS + system with Image LabTM Software (Bio-Rad Laboratories, CA, USA). Relative density of target proteins was determined with ImageJ software (NIH, MD, USA) and normalized to the GAPDH loading control.

After 6 days of decidualization, the medium from the cells was collected and the concentration of proteins prolactin and IGFBP-1 was measured using ELISA. To determine the concentration of prolactin, DiaMetra Prolactin ELISA Kit and protocol (DiaMetra S.r.l, Italy) was used. For IGFBP-1 concentration determination Mediagnost IGFBP-1 ELISA Kit was used (Mediagnost, Germany). The 96 plate with loaded samples was then measured using Infinite M200 Pro plate reader with i-control 1.5 software (Tecan, Männedorf, Switzerland). After optical density was detected at 450 and 630 nm wavelength, calibration curve was graphed based on the given standards of known concentrations. Then prolactin and IGFBP-1 concentrations (ng/ml) were calculated.

All experiments were repeated at least 3 times, unless specified otherwise. Statistical analysis was performed using GraphPad Prism (GraphPad Software, USA) software and one-way ANOVA test, followed by Tukey’s multiple pairwise comparison test. Statistical significance is marked as follows: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. Data were expressed as mean values ±SDs.

All measured endometrial and menstrual blood stem cells express mesenchymal cell surface markers, such as CD44 (homing cell adhesion molecule), CD73 (5′-nucleotidase), CD90 (thymocyte differentiation antigen 1) and CD105 (endoglin) and at a similar level as identified by flow cytometry (Figure 1). In most of the cases, more than 95% of cell population were marked as strongly positive for these typical mesenchymal stem cell markers. Less than 1% of EndSCs and MenSCs expressed CD34 (hematopoietic progenitor cell antigen), CD45 (leukocyte common antigen) and CD117 (c-kit) surface markers and they were characterized as negative. No significant differences between early and late passages of EndSCs and MenSCs were detected.

While analyzing perivascular marker’s CD146 expression, it was noticed that EndSCs expression was lower than that of MenSCs, although the expression between passages was similar. Both EndSCs and MenSCs expressed activated leukocyte cell adhesion surface marker CD166 at a high level at its expression was maintained at similar level throughout long-term cultivation.

The cells were also characterized by their proliferation rate. During continuous cultivation of EndSCs and MenSCs, the doubling time and viability of these cells was assessed. Examination of EndSCs showed that cell doubling time of early and late passages was very similar and around 2.5–3 ± 0.3 days. The doubling time of MenSCs was approximately 3 ± 0.2 days and did not differ significantly between early and late passages.

The ability of stem cells to maintain its stemness during numerous duplications was evaluated by gene expression of common pluripotency markers: OCT4, SOX2, NANOG and KLF4 using RT-qPCR. The appropriate Ct value of every gene and cell line is listed in the Supplementary Table S2. For this experiment as a control group human amniotic fluid mesenchymal stem cells (AF-MSCs) were used since they resemble pluripotent stem cells the most and are the closest to embryonic stem cells that we can use in laboratory setting.

Relative gene expression values of SOX2 gene in endometrial and menstrual blood stem cells were very similar in early and late passages, however, the values in most cases were lower than those of a control group—AF-MSCs (Figure 2A). During long-term cultivation of EndSCs and MenSCs OCT4 and NANOG gene expression was consistent, although it was noticed that the expression of these markers was decreased slightly in MenSCs compared to EndSCs, but these changes were not statistically significant (Figures 2B,C). Another mesenchymal stem cells gene-marker KLF4, responsible for regulation of stemness and proliferation, showed a similar expression in early as well as late passages in endometrial and menstrual blood stem cells (Figure 2D).

Next six genes (p53, p21, Rb, CCNA2, CCNE2, CDK2) associated with cell cycle regulation were analyzed. Sudden upregulation or downregulation of these genes could indicate that cells are reacting to possible negative environmental stressors. For example, upregulation of genes such as p53, p21, Rb is associated with cell cycle arrest, on the contrary, the downregulation of CCNA2, CCNE2, CDK2 genes will lead to decreased proliferative activity.

RT-qPCR analysis identified that p53 gene’s expression in endometrial and menstrual blood stem cells was at a comparable level to control group and did not fluctuate during long-term cultivation (Figure 3A). A relative gene expression of p21 gene was upregulated in EndSCs and MenSCs in late passages compared to early passages, however, such change was not significant (Figure 3B). The expression profile of Rb gene showed a similar tendency as p53 gene—expression remained at a similar level throughout long-term cultivation, but was slightly higher than in control cells (Figure 3C). By evaluating CCNE2 gene’s relative expression no significant changes were observed—expression values were at a similar level in early and late passages of EndSCs and MenSCs as well as in control group (Figure 3D). Minimal differences, though, not significant are seen in CCNA2 and CDK2 genes expression profiles. Relative gene expression of CCNA2 and CDK2 in endometrial mesenchymal stem cells late passages was downregulated compared to early passages (Figures 3E,F). However, menstrual blood stem cells have maintained this expression at a stable position in both early and late passages.

Since the decrease in proliferation rate can occur not only due to DNA damage responses and other environmental stressors, but also due to the initiation of senescence in cells, it is important to analyze such changes in cells that go through long-term cultivation. In some instances, young cells can experience premature senescence, which slows down their proliferation potency and weakens therapeutic abilities. For this reason, genes JUND, TOP2A, ATM and MYC that are often related to senescence were examined by RT-qPCR. Prior to gene expression analysis, we carried out β-galactosidase test on cells to determine whether EndSCs and MenSCs exhibit β-galactosidase activity, which is often a characteristic of senescent cells. We performed this test following (Savickiene et al., 2016) protocol on early and late passage endometrium and menstrual blood stem cells. The representative images are provided in Supplementary Figure S1. The results did not demonstrate any significant differences between passages associated with senescence: the morphology remained similar throughout passaging and there was minimal to none β-galactosidase activity. To make sure that gene expression pattern was not altered in EndSCs and MenSC, we then continued to perform gene expression analysis using senescence associated genes. We demonstrated that endometrial stem cells expressed JUND gene at a higher level than control group and menstrual blood stem cells, though such increase is not statistically significant (Figure 4A). JUND expression in MenSCs did not display significant differences among different passages—it was constant and resembled control group. Relative TOP2A gene expression in endometrial stem cells late passages was downregulated in comparison to EndSCs early passaged and became similar to early passages of menstrual blood stem cells (Figure 4B). There was a small decline in the expression levels of TOP2A gene in late passages of MenSCs, but overall in both experimental groups this expression remained similar to that of a control group. The ATM and MYC genes show similar changes in their gene expression profiles. The expression of ATM gene in EndSCs and MenSCs was almost identical to that of MYC gene (Figures 4C,D). During long-term cultivation senescence marker genes remained similar to or lower than control group, which might indicate that genes responsible for senescence are not yet activated in these cells.

Further, we explored with epigenetics and cell cycle regulation associated changes at protein expression levels in early and late passages of EndSCs and MenSCs. As demonstrated by Western blot analysis, levels of histone methyltransferases EZH2 (Enhancer of Zeste 2 Polycomb Repressive Complex 2 Subunit) and SUZ12 were downregulated in EndSCs and MenSCs during propagation and became lower than in control group cells (Figure 5A).

These two proteins are members of Polycomb repressive complex 2, which in stem cells is responsible for the maintenance of bivalent state of chromatin. Protein levels of modified histone H3K27me3, which is associated with transcriptional repression and inactive chromatin, remained the same in early and late passages of EndSCs, but were slightly enhanced in MenSCs late passages. Although, this alteration is statistically insignificant, but it may be linked to repression of certain genes. The next assessed protein group represents proteins that are involved in cell cycle control. Both cyclin E1 and p53 proteins’ levels fluctuated depending on a patient, but overall did not show any significant changes and were mostly maintained at a control group level (Figure 5A). Relative band intensity of detected proteins and protein levels were presented by graphical representation (Figure 5B). The assessment of protein levels of EZH2, SUZ12, and H3K27me3 along with gene expression profiles allow us to foresee changes that are happening, when cells duplicate multiple times. This can later be used to determine suitability of such cells to be used in therapy as potential regeneration tool.

As mentioned before, decidualization plays a critical role in implantation of blastocyst and eventually pregnancy, for this reason a large focus of this investigation was directed towards the assessment of decidualization during long-term cultivation. The cells were evaluated by a few criteria: morphology, decidualization gene markers’ PRL, IGFBP-1, WNT4 expression and secreted protein levels. Decidualization was induced by cultivating EndSCs and MenSCs in a medium containing db-cAMP and MPA for 3 and 6 days. For this experiment undifferentiated EndSCs and MenSCs were selected as control groups. EndSCs and MenSCs after 3 and 6 days of decidualization became rounder, flatter and reminiscent of epithelial cells compared to undifferentiated cells (Figure 6). As seen in the pictures, morphological alterations after 6 days were more prominent than after 3 days—more cells changed their phenotype from elongated to round, and most probably longer cultivation in conditioned medium has helped cells to achieve a more similar shape to that of epithelial cells. Moreover, there is no significant difference between early and late passages, which could mean that both groups are evenly capable to change their phenotype based on specific inductors. To further explore the process of decidualization on a molecular level, we evaluated gene expression levels of three main decidualization gene markers: PRL (prolactin), IGFBP-1 (Insulin-like growth factor-binding protein 1), WNT4 (Wnt Family Member 4). As determined by RT-qPCR, after decidualization induction relative gene expression of PRL, IGFBP-1, WNT4 was upregulated in all induced cells (Figure 7A). In endometrial and menstrual blood stem cells the expression of PRL gene remained at a similar level throughout propagation and after 3 and 6 days of decidualization. PRL expression in MenSCs decreased in comparison to EndSCs, although these changes were minor. EndSCs and MenSCs expressed IGFBP-1 in a similar manner as PRL gene: the expression was higher in EndSCs compared to MenSCs, but these differences were not significant. While analysing WNT4 gene expression profile it is seen that the expression of this gene was increased the most in early passages of EndSCs (Figure 7A). In all other groups WNT4 expression was maintained at a similar state during whole period of decidualization. It is important to mention, that longer decidualization of 6 days did not result in better expression of decidualization gene markers. Next, we measured levels of proteins secreted into medium during decidualization. Medium after decidualization induction was collected after 6 days and levels of PRL and IGFBP-1 proteins were assessed by ELISA. Both protein markers increased in comparison to undifferentiated control (Figure 7B). The secreted PRL levels were the highest in early passages of EndSCs, later it decreased slightly and became similar to MenSCs levels. Secreted IGFBP-1 protein levels in endometrial stem cells did not change during long-term cultivation.

Moreover, both EndSCs and MenSCs secreted the IGFBP-1 at a similar level and remained consistent throughout propagation. Overall, our findings suggest that in induced endometrial and menstrual blood stem cells the process of decidualization was initiated and late passages just as well as early passages maintain similar ability to undergo differentiation.