Latent Tri-lineage Potential of Human Menstrual Blood–Derived Mesenchymal Stromal Cells Revealed by Specific In Vitro Culture Conditions

Human menstrual blood–derived mesenchymal stromal cells (MenSCs) have become not only an important source of stromal cells for cell therapy but also a cellular source for neurologic disorders in vitro modeling. By using culture protocols originally developed in our laboratory, we show that MenSCs can be converted into floating neurospheres (NSs) using the Fast-N-Spheres medium for 24–72 h and can be transdifferentiated into functional dopaminergic-like (DALNs, ~ 26% TH + /DAT + flow cytometry) and cholinergic-like neurons (ChLNs, ~ 46% ChAT + /VAChT flow cytometry) which responded to dopamine- and acetylcholine-triggered neuronal Ca2+ inward stimuli when cultured with the NeuroForsk and the Cholinergic-N-Run medium, respectively in a timely fashion (i.e., 4–7 days). Here, we also report a direct transdifferentiation method to induce MenSCs into functional astrocyte-like cells (ALCs) by incubation of MenSCs in commercial Gibco® Astrocyte medium in 7 days. The MSC-derived ALCs (~ 59% GFAP + /S100β +) were found to respond to glutamate-induced Ca2+ inward stimuli. Altogether, these results show that MenSCs are a reliable source to obtain functional neurogenic cells to further investigate the neurobiology of neurologic disorders.


Introduction
Mesenchymal stromal cells (MSCs) are multipotent cells with self-renewal ability and differentiating potential [1,2]. Because of these meaningful features, MSCs hold great promise for the treatment and modeling of neurological disorders such as Alzheimer's (AD) and Parkinson's diseases (PD), and for testing potential therapeutic approaches (e.g., [3,4]). Since MSCs can be obtained from fetal (e.g., umbilical cord Wharton's jelly (UC-WJ), umbilical cord blood, placenta) and adult sources (e.g., dental pulp, gingival tissue, adipocyte tissue, menstrual blood) with minimal technical limitations, minor ethical issues, and/or reduced tumorigenic risk, they have emerged as a useful strategy for regenerative medicine [5]. Accordingly, menstrual blood stromal cells (hereafter named as MenSCs) have become not only an important source of stromal cells for cell therapy (e.g., [6,7]) but also as a cellular source for neurologic disorders in vitro modeling. Indeed, it has been shown that MenSCs can be converted into clonogenic neurosphere-like cells (NSCs) in 10-20 days, which can be transdifferentiated into glial-like and neural-like cells in 12-16 days [8,9]. However, no further data are available to determine whether MenSCs can transdifferentiate into other neuronal lineages such as dopaminergic, cholinergic, or astrocytic neuronal cells [10].
Recently, we have derived MSCs from UC-WJ to obtain neurospheres (NSs), i.e., free-floating spherical cell aggregates, also named as neural precursor cells, in 24 h, and NS-derived nerve-like cells (NLCs) in 4 days [11]. Moreover, MSCs have also been transdifferentiated into functional Diana Quintero-Espinosa andViviana Soto-Mercado contributed equally to this work cholinergic-like neurons (ChLNs) in 7 days [12]. While UC-WJ-derived MSCs are available for only a minority of individuals who have their samples banked at birth, MenSCs can be isolated with minimal invasiveness and risk to the donor and can be obtained in sufficient numbers to enable expansion in culture. Therefore, MenMSC might represent a fast, safe, and efficient way of generating neural-like cells [4].
The present study aimed to investigate whether MenSCs can transdifferentiate into NSs and neuronal lineages with culture protocols originally developed in our laboratory including three culture media such as the Fast-N-Spheres [11], the NeuroForsk [11], and the Cholinergic-N-Run medium [12]. Similar to UC-WJ-MSCs, we show that Men-SCs can be converted into floating NSs using the Fast-N-Spheres medium for 24-72 h and can be transdifferentiated into functional dopaminergic-like neurons (DALNs) and ChLNs which responded to dopamine-and acetylcholinetriggered neuronal Ca 2+ inward stimuli when cultured with the NeuroForsk and the Cholinergic-N-Run medium, respectively, in 4-7 days of culture. We also report a direct transdifferentiation method to induce MenSCs into functional astrocyte-like cells (ALCs) by incubation of MenSCs in commercial Gibco® Astrocyte Medium in 7 days. Accordingly, MenSC-derived ALCs (~ 59% GFAP + /S100β +) were found to respond to glutamate-induced Ca 2+ inward stimuli. Altogether these results show that MenSCs are a reliable source to obtain functional neurogenic cells to further investigate the neurobiology of AD and PD neurologic disorders.

Isolation and Characterization of Mesenchymal Stromal Cells Derived from Human Menstrual Blood (MenSCs)
The menstrual blood samples were collected from three healthy females aged between 18 and 30 years. Donors provided a signed informed consent approved by the ethics committee of the Sede de Investigación Universitaria (SIU), University of Antioquia, Medellín, Colombia. Menstrual blood (MenB) was collected by cup collection (10-15 mL) during the first 3 days of menses. Briefly, menstrual blood samples were delivered into the laboratory and mixed with an equal volume of phosphatebuffered saline (PBS) containing 1 mM ethylenediamine tetra-acetic acid (EDTA), with 100 U/mL penicillin/streptomycin 0.25 mg/mL amphotericin B, and subject to cell lysis or standard Ficoll procedures within 24 h as previously described in ref. [13]. After centrifugation, the cells were suspended in a buffy coat (6.7 × 10 6 ± 3 × 10 6 cells, n = 3), were transferred into a new tube, washed in PBS twice, re-suspended in growth medium (low-glucose DMEM medium supplemented with 10% FBS (Gibco, USA), 100 U/mL penicillin/streptomycin, and 0.25 mg/mL amphotericin B), and seeded into 25-cm 2 plastic cell culture flasks at 37 °C with 5% humidified CO 2 . The medium was replaced every 3 days leaving behind the adherent cells that were growing as fibroblastic cells in clusters. When the cells reached 80-90% confluence (P0), the cells were detached by 0.25% trypsin/1 mM EDTA and subcultured to new flasks by the ratio of 1:3. The isolated MenSCs were evaluated for the following characteristics: colony formation capacity (i.e., colonies with a typical adherent growth, colony-forming unit fibroblast activity, and spindle-shaped and fibroblast-like morphology); positivity for the mesenchymal-associated surface markers CD90, CD9, and CD73 (99% positive); and karyotype with normal shape, number, and distribution. The differentiation capacity in an osteoblast, chondrocyte, and adipocyte lineage of MenSCs, as well as the presence of neuronal precursors and astrocyte markers (e.g., NFL, β-TUB III, GFAP, S100β, TH, ChAT), was assessed according to refs. [12,14].

Karyotyping
Karyotype analysis was performed by Genetica Lab, S.A.S., using standard cytogenetic protocols. Briefly, at 60-70% confluence, MenSCs cells were incubated with 0.1 mg/ mL Colcemid (Sigma) for 90 min at 37 °C. Then, the cells were detached with 0.25% trypsin and centrifuged at 591 × g for 20 min. The medium was removed, and the hypotonic solution (0.075 M KCl, 0.017 M Na-citrate) was added and incubated for 20 min at 37 °C. After a new centrifugation, cells were fixed with freshly prepared Carnoy's solution. Metaphase spreads were analyzed after staining with quinacrine (Sigma) for karyotyping. Analysis was performed on three different primary cultures counting 20 metaphases for each sample.

Immunofluorescence (IMF) Analysis
For immunofluorescence analysis of neural and astrocytes markers, cells treated with MCm, NeuroForsk, Gibco® Astrocyte Medium [15][16][17] or Ch-N-Rm medium for 0, 4, and 7 days and Neurospheres Fast medium for 0, 24, and 72 h were fixed with paraformaldehyde for 20 min, followed by Triton X-100 (0.1%) permeabilization and 5% bovine serum albumin (BSA) blockage. Cells were then incubated overnight with primary antibodies against DAT, TH, GFAP, S100β, β-TUB III, MAP2, NFL, VAChT, and ChAT proteins (1:500). After exhaustive rinsing, we incubated the cells with secondary fluorescent antibodies (DyLight 488 and 595 donkey anti-rabbit, -goat, and -mouse, Cat DI 2488 and DI 1094, respectively) 1:500. The nuclei were stained with Hoechst 33342 (1 μM, life technologies) and images were acquired on a Floyd Cells Imaging Station microscope. Imaging processing data for further statistical analysis were gathered by ImageJ software (http:// imagej. nih. gov/ ij/). The figures were transformed into 8-bit images and the background was subtracted. The cellular measurement regions of interest (ROI n = 7 from three different experiments) were drawn around nuclear or over all cells (for cytoplasmic probes) and the fluorescence intensity was subsequently determined applying the same threshold. For mean fluorescence intensity (MFI) was obtained by normalizing total fluorescence to the number of nuclei.
Cells treated with MCm, NeuroForsk, astrocyte, or Ch-N-Rm medium at days 0, 4, and 7 were detached with 0.25% trypsin-EDTA 1 mM and fixed in suspension with paraformaldehyde (overnight). After washing, cells were simultaneously incubated with GFAP, S100β, DAT, TH, ChAT, and VAChT primary antibodies (1:500) at 4 °C overnight. Cell suspensions were washed and incubated with DyeLight 594 donkey anti-goat and DyeLight 488 donkey anti-rabbit antibodies (1:500). Finally, cells were washed and re-suspended in PBS for analysis on a Canto cytometer (Beckman Coulter). Ten thousand events were acquired and the acquisition analysis was performed using FlowJo 7.6.2 Data Analysis Software. Positive staining was defined as the fluorescence emission that exceeded levels of the population stained with the negative control (only secondary antibodies staining).

Data Analysis
Statistical analyses were conducted using the Student t analysis or one-way ANOVA followed by Bonferroni post hoc comparison calculated with the GraphPad Prism 6 Scientific Software (GraphPad, Software, Inc. La Jolla, CA, U.S.A.). Statistical significance was accepted at *p < 0.05, **p < 0.01, and ***p < 0.001.
Next, we wanted to evaluate the presence of neural and astrocyte-associated molecules in MenSCs under in minimal culture medium (MCm). Analysis of protein expression by Western blotting showed a basal level of NFL ( Fig. 2A, B), β-TUB III ( Fig

In Vitro Transdifferentiation of MenSCs into Dopaminergic-Like Neurons (DALNs) Involves the Upregulation of Specific Neural and Dopaminergic Proteins
To assess the ability of MenSCs to transdifferentiate into DALNs, MSCs were cultured with NeuroForsk medium for 4 or 7 days [14]. To further confirm the dopaminergic lineage of MenSCinduced DALNs, we analyzed TH and DAT double staining. As shown in Fig. 4, flow cytometry (FC) analysis shows a statistically significant increase in the percentage of TH/ DAT expressing cells at day 4 (~ 14%) and 7 (~ 26%) of cellcultured in NeuroForsk medium compared to MenSCs at day 0 (~ 8%, Fig. 4A, B). We found no statistical difference

Dopamine Triggers Intracellular Ca 2+ Accumulation in MenSC-Derived DALNs
Based on the significant increase of dopaminergic cellular populations in MenSCs exposed to the NeuroForsk medium, we evaluated the functional response of DALNs under a dopamine stimulus. Therefore, the cytoplasmic Ca 2+ accumulation in DALNs was evaluated with Fluo-3-mediated Ca 2+ imaging. While DALNs were unaffected by dopamine at day 0, dopamine-induced a transient intracellular Ca 2+ elevation at days 4 and 7 (Fig. 4C, D) under the same experimental conditions. Furthermore, the maximal fluorescence change (ΔF/F) at day 4 was 0.75 ± 0.05-fold after 10 s of dopamine exposure compared to cells at day 0 (p < 0.05), whereas the maximal ΔF/F at day 7 was 0.25 ± 0.2-fold after 20 s of dopamine addition (p < 0.05) and it remains stable until the 50 s (Fig. 4D).
Since the choline acetyltransferase (ChAT) enzyme and vesicular acetylcholine transporter (VAChT) protein are specific cholinergic cell lineage markers [22,23], we wanted to confirm whether the MenSCs transdifferentiated into ChLNs simultaneously expressed ChAT and VAChT. Accordingly, we establish the percentage of MenSCs derived ChLNs with Cholinergic-N-Run medium and we assessed the proportion (%) of ChAT/VAChT immunopositive neurons after the cholinergic induction. Flow cytometry analysis shows a statistically significant increase in the percentage of ChAT/ VAChT expressing cells at day 4 (~ 39%) and day 7 (~ 46%) of culture exposure compared to MenSCs at day 0 (~ 28%) (Fig. 6A, B). We found no statistical difference (p < 0.05) in the level of double expression of ChAT/VAChT in ChLNs at 4 and 7 days of culture.

Acetylcholine Triggers Intracellular Ca 2+ Accumulation in MenSC-Derived ChLNs
The above findings encouraged us to evaluate the response of ChLNs cells to acetylcholine culture in Cholinergic-N-Run medium. Therefore, the cytoplasmic Ca 2+ accumulation in ChLNs was evaluated with Fluo-3-mediated Ca 2+ imaging. While ChLNs were unaffected by ACh at day 0 of Cholinergic-N-Run medium culture, ACh induced a transient intracellular Ca 2+ elevation at day 4 and day 7 ( Fig. 6C, D) under the same experimental conditions. Furthermore, the maximal fluorescence change (ΔF/F) at day 4 was 6.14 ± 1.5-fold and the maximal ΔF/F at day 7 was 5.57 ± 1.2-fold after 10-s ACh exposure compared to cells at day 0 (p < 0.05, Fig. 6D).

Transdifferentiation of MenSCs into Astrocyte-Like Cells (ALCs) Involves the Upregulation of Specific Glial Proteins
The transdifferentiation of MenSCs into astrocytes was evaluated after 0, 4, and 7 days of incubation with astrocyte medium®. As shown in Fig. 7, MenSC-derived ALCs express statistically significant increased expression level of S100β (e.g., ~ 1.2 f.i., Fig. 7A, C) and GFAP (~ 0.8 and ~ 0.6 f.i.,  (Fig. 7C, D). Western blotting analysis shows that MenSC-derived ALCs express almost basal level of protein NFL (Fig. 7A, E) and β-TUB III (Fig. 7A, F). Similar results were obtained by IMF analyses (Fig. 7G-P). Since GFAP is the gold standard marker to identify astrocytes and S100β indicates a mature stage of astrocytes, we evaluate those markers in the MenSC-derived ALC population. Accordingly, FC analysis shows a statistically significant increase in GFAP/S100β double-positive cells in ALCs at day 4 (~ 25%) and 7 (~ 59%) of culture compared to cells cultured at day 0 (Fig. 8A, B).

Glutamate Triggers Intracellular Ca 2+ Accumulation in ALCs
Intracellular astrocyte calcium signaling is triggered by multiple factors, including glutamate at the synaptic cleft, which stimulates G protein-coupled receptors on the membrane, leading to the production of IP3. IP3 activates IP3R2 receptors on the endoplasmic reticulum (ER), resulting in intracellular calcium release and subsequent exocytosis [24]. Therefore, the cytoplasmic Ca 2+ accumulation in ALCs was evaluated with Fluo-3-mediated Ca 2+ imaging. While ALCs were unaffected by glutamate at day 0 of Astrocyte medium culture, glutamate-induced a transient intracellular Ca2 + elevation at day 4 and day 7 (Fig. 8C, D) under the same experimental conditions. Furthermore, the maximal fluorescence change (ΔF/F) at day 4 was 0.10 ± 0.05-fold after 120 s of glutamate exposure compared to cells at day 0 (p < 0.05), whereas the maximal ΔF/F at day 7 was 0.50 ± 0.2-fold after 240 s of glutamine addition (p < 0.05) and decline slowly.

In Vitro Transdifferentiation of MenSCs into Neurospheres/Spheroids
To determine whether MenSCs transdifferentiate into NSs, we initially determined the morphology and the presence of molecules associated with NSs formation in cells treated with Fast-N-Spheres medium after 0, 1, and 3 days of incubation according to a previous report [11]. As shown in Fig. 9, MenSCs cultured with this specific display the typical morphology of NSs as early as 1 day and formed large NSs at 3 days when compared to 0 days of incubation. Interestingly, both the MenSCs (0 days) and the formed NSs (at 1 or 3 days) expressed high levels of SOX2 and NESTIN proteins according to Western blot analysis (Fig. 9A-C). These results were confirmed by IMF analysis (Fig. 9D-K).

Discussion
The stromal cell-based disease model provides a platform for a better understanding of human neurodegenerative disease mechanisms (e.g., [25]) and the potential discovery of innovative therapeutics (e.g., [26]). As primary human neurons from living subjects are normally not accessible to researchers, there is a pressing need for an alternative source of authentic human neurons which allows modeling of neurodegeneration in vitro. Therefore, optimal cell culture conditions and timing are critical for experimental success. Recently, our laboratory has developed three original culture media known as NeuroForsk, Cholinergic-N-Run, and Fast-N-Spheres media to obtain dopaminergic-like (DALNs), cholinergic-like (ChLNs) neurons, and neurospheres (NSs) from UC-WJ-MSCs [11,12]. By using those culture media, which seems more efficient compared to the already used protocol in developing neural-like cells, we have successfully transdifferentiated MenSCs into functional DALNs, ChLNs, and NSs in 4-7 days of cell culture. Likewise, by using a commercial culture media (e.g., Astrocyte media®), we also obtained functional astrocyte-like cells (ALCs) in a similar time frame (i.e., 4-7 days). Also, MenSCs can differentiate into various mesoderm cell lineages, such as adipocyte (this work), chondrocytes [27], and osteocytes [28,29] comparable to MSCs obtained from several tissue sources, including perinatal (e.g., WJ-MSCs [30,31]), bone marrow, adipocyte, dental pulp, and human efflux (e.g., MenSCs [7,10]), among others. Furthermore, naïve Men-SCs (day 0) express basal protein levels of neuronal markers NFL, β-TUB III, TH, DAT, ChAT, VAChT, and MAP2. In contrast to Azedi et al., 2014 [9], we found that MenSCs express GFAP. Moreover, the expression of S100β protein confirms that MenSCs are primed to display glial-specific imaged, N = 3 dishes) as a response to ACh treatment. ACh was puffed into the culture at 0 s (arrow). Then, the Ca 2+ fluorescence of cells was monitored at indicated times. Color contrast indicates fluorescence intensity: dark blue < light blue < green < yellow < red. (D) Normalized mean fluorescence signal (ΔF/F) over time, indicating temporal cytoplasmic Ca 2+ elevation in response to ACh treatment. One-way ANOVA, post hoc test Bonferroni. Data are presented as mean ± SD (*p < 0.05; **p < 0.01; ***p < 0.001) lineage markers. Differential experimental procedures might explain those divergent results. Taken together, these observations suggest that not only MenSCs differentiate in mesoderm lineages, but also express neuronal precursors typical of ectoderm lineages (i.e., neural lineage). These cellular features together with the fact that MenSCs have no ethical concerns, are easily recovering from women, are unexpensive and time-saving, and make MenSCs ideal for disease modeling.
We report for the first time that MenSCs can transdifferentiate into DALNs (~ 26%) cultured in NeuroForsk media for 7 days. Three observations support these findings. First, transdifferentiated DALNs expressed high levels of specific dopaminergic lineage markers TH-the enzyme necessary for the dopamine production from its precursor L-tyrosineand DAT-a dopamine transporter plasma membrane glycoprotein, as well as expressed high levels of neuronal-specific protein markers NFL, MAP2, and β-TUB III. Both DAT and TH expression identifies bona fide dopaminergic neurons [32]. Second, the percentage rates of protein expression of TH (~ 29% + cells), DAT (~ 33% + cells), and (~ 26% TH/ DAT double-positive cells) found in DANLs are comparable or even higher than previously reported from other sources such as those obtained in WJ-MSC cells after 7 days of differentiation with NeuroForsk only (~ 25% TH + [14]), or bone marrow-MSC (~ 26% TH + ; ~ 16% DAT +), the dental pulp (~ 29% TH + ; ~ 23% DAT +), and adipocyte-MSC (~ 29% TH + ; ~ 23% DAT +) after 14 days of differentiation with Forskolin and FGF2 [33]. Finally, dopamine signaling evoked intracellular Ca 2+ concentration changes in DALNs. Interestingly, the expression of TH (~ 17%) and DAT (~ 10%) proteins in naïve MenSCs reveal a strong potential of MSCs to generate DALNs. Moreover, although we found 8% TH/ DAT double-positive cell markers at day 0, undifferentiated cells do no exhibit a functional phenotype in contrast to DALNs at days 4 and 7 (Fig. 4). This observation can be explained most probably by the absence of dopaminergic D 2 -receptors in naïve MenSCs. However, mature Bonferroni. Data are presented as mean ± SD (*p < 0.05; **p < 0.01; ***p < 0.001) dopaminergic neurons display an increase of dopaminergic receptors that are linked to synaptogenesis processes and Ca 2+ influx [34].
Here, we also report for the first time that MenSCs transdifferentiate into ChLNs in 4 − 7 days. Effectively, MenSCderived ChLNs significantly express cholinergic markers ChAT-the enzyme that catalyzes the biosynthesis of ACh [35]-and VAChT-the ACh neurotransmitter transporter [22], two unique markers for cholinergic neurons according to WB, IMF, and FC analysis. In agreement with others, who have demonstrated that MSCs from dental pulp [36], adipose tissue [37], and UC-WJ [12] can transdifferentiate into ChNs/ChLNs, we show that MenSCs readily transdifferentiated into functional ChLNs. Similar to UC-WJ-MSCderived ChLNs [12], the MenSC-derived ChLNs are responsive to ACh stimuli on days 4 and 7. Interestingly, although naïve MenSCs express both cholinergic ChAT/VAChT and ACh receptors [38], they are unresponsive to ACh at day 0.
These observations suggest that the presence/expression of cholinergic markers in undifferentiated MenSCs (i.e., ChAT/ VAChT/AChRs) is not yet functional indicating a cellular immature state that necessitates an adequate micro-environmental stimulus, as provided by Cholinergic-N-Run media, to express functional (responsive) AChRs [39]. Interestingly, ChLNs maintained the expression of general neural lineage markers (e.g., NFL, β-TUB III) whereas other specific cell lineages markers diminished (e.g., GFAP, TH) at day 7 of transdifferentiating. Therefore, we conclude that MenSCderived ChLNs are mature neural cells obtained timelessly.
Neurospheres (NSs) are 3D structures composed of free-floating conglomerates of neural progenitor cells (NPCs) derived from either isolated primary tissue (e.g., embryonic tissue) or from induced pluripotent stem cells (iPSCs) and grown for limited periods. Therefore, NSs provide an unmatched platform to evaluate the stem-/ stromal-cell-like behavior of neurogenic tissue and can be used in a variety of in vitro experiments to delineate the molecular and cellular characteristics of neuronal cells for in vivo transplantation in neurodegenerative disorders (e.g., [48]). We have reported a fast and easy method to obtain NSs from UC-MSCs in 24 h by using the Fast-N-Spheres medium [11]. Under similar experimental conditions, we have been able to transdifferentiate MenSCs into NSs phenotypically similar to MSC-derived NSs. However, while the MenSC-derived NSs were NES-TIN-and SOX-2-positive-a key transcription factor in the regulation of pluripotency and neural differentiation [49]-the MSC-derived NSs were NESTIN-positive but SOX-2-negative [11]. These observations suggest that NSs derived from MenB conserved those specific markers similar to those observed in NSs from neurogenic niches in cerebral tissue [50]. Therefore, MenSC-derived NSs are a more reliable 3D cellular source for neurodevelopmental studies. Moreover, in agreement with previous data [8], the MenSC-derived NSs cultured for 3 days significantly increased the expression of neuronal (e.g., β-TUB III) as well astrocytic (e.g., GFAP) markers. We report for the first time that MenSC-derived NSs also express the specific neuronal lineage dopaminergic marker TH and cholinergic marker ChAT. Taken together, these results suggest that MenSC-derived NSs can highly regulate trineuronal lineage proteins typical of the dopaminergic, cholinergic, and glial neuronal cells, which might represent an excellent strategy not only to studying the tissular pathology of AD and PD but also to be used as starting material for the development of more complex structures such as organoids [51]. Therefore, as depicted in Fig. 10, MenSCs are a unique and reliable biological source for derivation of the most important neural cells such as dopaminergic, cholinergic, and astrocytic neuronal cells implicated in neurodegeneration (e.g., Alzheimer's and Parkinson's diseases).