ISSN 1990519X, Cell and Tissue Biology, 2013, Vol. 7, No. 3, pp. 235–244. © Pleiades Publishing, Ltd., 2013. Original Russian Text © V.I. Zemelko, I.B. Kozhukharova, L.L. Alekseenko, A.P. Domnina, G.F. Reshetnikova, M.V. Puzanov, R.I. Dmitrieva, T.M. Grinchuk, N.N. Nikolsky, S.V. Anisimov, 2013, published in Tsitologiya, Vol. 55, No. 2, 2013, pp. 101–110.
Neurogenic Potential of Human Mesenchymal Stem Cells Isolated from Bone Marrow, Adipose Tissue and Endometrium: a Comparative Study V. I. Zemelkob, I. B. Kozhukharovaa, b, L. L. Alekseenkob, A. P. Domninab, G. F. Reshetnikovaa, b, M. V. Puzanova, R. I. Dmitrievaa, T. M. Grinchuka, b, N. N. Nikolskyb, and S. V. Anisimova, b a
Almazov Federal Center for Heart, Blood and Endocrinology, St. Petersburg b Institute of Cytology RAS, St. Petersburg email:
[email protected] Received October 19, 2012
Abstract—Mesenchymal stem cells (MSCs) can be isolated from many adult tissue sources. These cells are a valuable substrate in cell therapy for a substantial number of diseases and injuries. Different types of MSCs vary in plasticity. We performed a comparative study of the neurogenic potential of three types of human MSCs derived from bone marrow (BMSCs), subcutaneous adipose tissue (ADSCs) and endometrium (iso lated from the menstrual blood) (eMSCs). It was shown that all three types of MSC cultures demonstrate multipotent plasticity and predisposition to neurogenesis, based on the expression of pluripotency marker SSEA4 and neuronal precursors markers nestin and betaIIItubulin. Further analysis revealed a transcrip tion of the neuronal marker MAP2 and neurotrophin3 in the undifferentiated BMSCs and ADSCs. Addi tionally, a significant basal level of synthesis of brainderived neurotrophic factor (BDNF) in the eMSC cul ture was also observed. Stimulation of neural induction with agents such as 5azacytidine, recombinant human basic fibroblast growth factor (bFGF), recombinant human epidermal growth factor (EGF), a recombinant human fibroblast growth factor 8 (FGF8), morphogen SHH (sonic hedgehog), retinoic acid (RA) and isobutylmethylxanthine (IBMX), showed further differences in the neurogenic potential of the MSCs. The components of the extracellular matrix, such as Matrigel and laminin, were also the important inducers of differentiation. The most effective neural induction in the BMSCs proceeded without the RA par ticipation while pretreated with 5azacytidine. In contrary, in case of eMSCs RA was a necessary agent of neural differentiation as it stimulated the transcription of neurotrophin4 and the elevation of secretion level of BDNF. The use of laminin as the substrate in the derived eMSCs appeared to be critical, though an incubation of the cells with 5azacytidine was optional. As far as the derived ADSCs, RA in combination with 5azacytidine caused the elevation of expression of MAP2, but reduced the secretion of BDNF. Thus, the effect of RA on neural differentiation of ADSCs is ambiguous and, together with the study of its signaling pathways in the MSCs, requires further research. The therapeutic effect of transplanted MSCs is commonly explained by their paracrine activity. The high basal level of BDNF synthesis in the eMSCs, along with their high proliferative rate, noninvasive extraction and neural predisposition, is a powerful argument for the use of the intact eMSCs as a substrate in cell therapy to repair a nerve tissue. Keywords: cell therapy, mesenchymal stem cells DOI: 10.1134/S1990519X13030140 1
Stem and progenitor cells in vivo perform impor tant biological functions, defining the processes of growth, development, adaptation and regeneration. In this matter, the dysfunction of circulating resident progenitor and stem cells, developed as a result of genetic, metabolic, and other issues, causes the delay
1
The article was translated by the authors. Abbreviations: MSCs—mesenchymal stem cells, ADT—subcu taneous adipose tissue, ADSCs—MSCs derived from ADT, BM—red bone marrow, BMSCs—MSCs isolated from BM, HSCs—hematopoietic stem cells, eMSCs—endometrial mes enchymal stem cells derived from menstrual blood, ESCs— embryonic stem cells.
of growth and development and the decrease of an adaptive and regenerative potential of an organism (Molero, 2009; Thornell, 2009). The most extensive niche of adult stem cells is the red bone marrow (BM) which comprises hematopoietic (bloodforming) stem cells (HSCs) and mesenchymal stem cells (MSCs). The MSCs from BM and other tissues perform a range of functions, providing trophic support for other cell types (e.g for BM it is HSCs), synthesizing and secret ing growth factors and components of signaling path ways, while possessing a high level of plasticity (Prockop, 2009; Schraufstatter, 2011).
235
236
ZEMELKO et al.
Currently, adult MSCs are widely used as an object of investigations of the mechanisms of cell prolifera tion, differentiation, and cell–cell interactions. In addition, the treatment of various diseases and injuries using the MSCs or their derivatives, based on the prin ciples of cell therapy, became a very important issue. On set of their properties, the MSCs have a number of advantages over all other types of stem cells. Although the proliferative potential of the MSCs is significantly lower than for embryonic stem cells (ESCs), induced pluripotent stem cells and HSCs, they are genetically more stable, more available for isolation and use, without encountering to ethical problems (Anisimov et al., 2009). In the last decade it has been found that the level of plasticity of the MSCs is much higher than previously thought: they are able to differentiate not only in numerous types of cells of mesenchymal origin, including cells of the bone, cartilage, muscle, tendon and fat, but also in other cell types, including neurons and glial cells (Woodbury et al., 2000). This has been demonstrated for MSCs from animal origin as well as for human MSCs. These properties of the MSCs dem onstrate the potential of use of these stem cells as the substrate for cell therapy, including treatment of neu rodegenerative disorders. Although MSCs are relatively few in BM (0.01– 0.001% of total nucleated cells), technology of the MSCs selection from BM aspirate presents no techni cal difficulties. However, the procedure of BM aspira tion is associated with a number of complications. It is difficult in certain clinical situations, and the mobili zation of stem cells of BM in the peripheral blood with pharmaceutical agents (granulocyte colony stimulat ing factor Neupogen) is very costly. Therefore, the possibility to use alternative sources of MSCs other then BM for cell therapy purposes becomes greatly favorable. Recently MSCs were isolated from a wide range of adult tissues, including subcutaneous visceral adipose tissue, skeletal muscle tissue, cartilage, tendons, den tal pulp, endometrium, periodontal ligament, synovial membrane and lungs. It has been suggested that the population of resident MSCs exist in almost all organs and tissues of adults (da Silva Meirelles et al., 2006), while MSCs of different origin have some differences on a number of parameters (Dmitrieva et al., 2012). Nowadays sabcutaneous adipose tissue and menstrual blood (containing endometrial cells) are considered as the most accessible sources for MSCs isolation. In this work we performed a comparative analysis of the neurogenic potential of human MSCs derived from bone marrow (BMSCs), subcutaneous adipose tissue (ADSCs) and endometrium (isolated from the menstrual blood) (eMSCs).
MATERIALS AND METHODS Aspiration procedure of BM and ADT samples per formed at the Federal Center of Heart, Blood and Endocrinology of V.A. Almazov (St. Petersburg), menstrual blood specimen collection made at the International Centre of Reproductive Medicine (St. Petersburg). The study was conducted in accor dance with the standards of the Helsinki Declaration (1989). Aseptic BM aspiration was performed under a local anesthesia by the aspiration needle (13 G or 14 G) form the iliac or handles sternum in a single puncture; total BM aspirate volume was 3.0–6.0 mL, anticoagu lation was performed in ready vials (BD Biosciences, USA) containing dipotassium edetate (K2EDTA) at a concentration of 1.8 mg K2EDTA per mL of BM. The samples were transported to the laboratory for MSCs isolation at +4°C within 4 hours after collec tion. Aseptic ADT aspiration was performed under a local anesthesia from the left umbilical region by the thick needle (19 G) in a single puncture; total ADT aspiration was 0.3–0.5 mL. Then the samples were transported to the laboratory for MSCs isolation at +4°C within 2 hours after collection. Intake of a menstrual blood was carried out on the 2nd day of the menstrual cycle. The blood was col lected by aspirator Ipas MVA plus with 4 mm in diam eter nozzle, introduced into the cervical canal. An average volume of collected blood from each patient was 1.0–2.0 mL. The samples was transferred to a phosphate–buffered saline (PBS) (Sigma, USA) with addition of sodium citrate and 1% antibiotic and anti mycotic mixture (mixture of penicillin, streptomycin and amphotericin B) (Sigma, USA). In this solution the samples of menstrual blood were delivered to the laboratory for MSCs isolation within 2–4 hours at +4°C. Primary cultures of the MSCs. The BM aspirate samples were diluted 4 times with sterile PBS. Then mononuclear cells of BM were isolated by density gra dient separation, using reagent FicollPaque PLUS (Amersham, USA). The isolated mononuclear cells (10–20 million) were seeded in culture flasks and cul tivated under standard conditions (5% CO2, 37°C), first (1–2 passages) in the culture medium of alpha MEM (Biolot, Russia), and then in DMEM/F12 (Invitrogen, USA) containing 10% optimized for MSCs fetal bovine serum (FBS; HyClone, USA), 1% of a mixture of antibiotics (penicillin and streptomy cin) and 1% of Glutamax (Invitrogen, USA). Non adherent cells were washed after 48 hours of cultiva tion. As the derived BMSCs reached the monolayer, they were split using 0.05% trypsin and EDTA (Invit rogen, USA). The ADT aspirate samples were washed from visi ble blood clots with sterile Hanks balanced salt solu tion (HBSS, Biolot, Russia) and mechanically disso CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
NEUROGENIC POTENTIAL OF HUMAN MESENCHYMAL STEM CELLS
ciated. The enzymatic dissociation was performed using a sterile solution of collagenase III (5 mg/mL) (Worthington, USA) on a sterile HBSS for 20– 30 minutes. After neutralization of the collagenase with 10% FBS, all dissociated cells were transplanted into culture flasks and cultured as described above for the BMSCs. The solutions of the menstrual blood samples were centrifuged. The pellets containing fragments of endometrium were resuspended in PBS, containing 10% of a mixture of antibiotic and antimycotic, and incubated for 1 h at 37°C. The recentrifuged cells were transferred into T25 culture flasks (Fisher Scien tific, USA) and cultivated in DMEM/F12, supple mented with 10% FBS, 1% antibiotic and antimycotic mixture and 1% Glutamax. To obtain an adhesive cell population, the cells were cultured under standard conditions for 3–7 days while the medium was changed several times. The resulting adhesive cell populations (4–9 million) were subcultured using 0.05% trypsin and EDTA solution, twice a week at 1 : 3 ratio. Flow cytometry. A single cell suspension of the MSCs was obtained by using 0.05% trypsin and EDTA solution. 1 × 106 cells were suspended in 1 mL of PBS, containing 5% FBS. For the MSCs immunopheno typing antibodies to CD9, CD14, CD34, CD44, CD45, CD90, CD11a, CD13, CD19, CD29, CD73, CD105, CD117, CD146, HLADR, conjugated to PE or FITC or APC (BectonDickinson BioSciences, USA),were used. Immunophenotypic analysis of the MSCs on CDsurface markers was performed using a flow cytometers Epics XL (Beckman Coulter, USA) and BD FACS Calibur using software CELLQuest. Neural differentiation of the MSCs in vitro. The MSC cells were dissociated with 0.05% trypsin and EDTA and subcultured in 35 mm Petri dishes (1 × 105 cells) and in 60 mm Petri dishes (3 × 105 cells) with coverslips coated with Matrigel (Matrigel; BD Bio science, USA) or laminin (50 μg/mL) (Sigma, USA). To induce neural differentiation two protocols were implemented. According to Protocol 1 on the following day the growth medium in the dishes was changed to serum free differentiation medium consisting of DMEM/F12, 1% antibiotic/antimycotic mixture, 1% glutamax, 1% reagent B27 (B27 supplement; Gibco, USA), 1% reagent N2 (N2 supplement; Gibco, USA), 25ng/mL of human recombinant basic fibroblast growth factor (bFGF; BD Bioscience, USA) and 25 ng/mL of human recombinant epidermal growth factor (EGF; CALBIOCHEM, USA). After 2 days without changing the medium, the following reagents were added: 1 μM of retinoic acid (RA; Sigma, USA) and 0.25 mM of isobutylmethylxanthine (IBMX; Sigma, USA) for 1–2 days for the BMSCs and the ADSCs and for 4 days for the eMSCs. To identify the most efficient induction of the neural differentiation for the different MSCs, experiments were performed CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
237
with two modifications: 1) preincubation of the cells with 10 μg/mL of 5azacytidine added into the growth medium for 2 days and 2) without the addition of RA. According to Protocol 2 on the following day the growth medium in the Petri dishes was changed to serumfree differentiation medium consisting of DMEM/F12, 1% antibiotic/antimycotic mixture, 1% glutamax, 1% of reagent B27, 1% of reagent N2, 250 ng/mL of SHH (sonic hedgehog, R & D, USA), 50 ng/mL of bFGF and 100 ng/mL of human recom binant fibroblast growth factor 8 (FGF8; R & D, USA). After 2 days of incubation without changing the medium, the following reagents were added: 1 μM of RA and 0.25 mM of IBMX for 1–2 days for the BMSCs and for 4 days for the eMSCs. Immunocytochemistry. Fluorescent staining of antigens nestin, betaIIItubulin, neuronal intermedi ate filaments heavy (NFH), neuronal nuclei (NeuN), glial fibrillary acidic protein (GFAP) and microtubules associated protein MAP2 was performed according to standard protocol. The cells were fixed with 4% formalin solution and permeabilized with 0.1% solution of Triton X100 (Merck, USA) in PBS. Nonspecific binding of the antibodies was blocked using 1% bovine serum albumin (BSA; Sigma, USA) and 5% solution of goat serum (Sigma, USA) both in PBS for 30 min. The antibodies used were mouse monoclonal against betaIIItubulin in a dilution of 1 : 1000 (Chemicon, USA), NFH (Abcam, USA), Neun (Millipore, USA), MAP2 (Millipore, USA) all in a ratio of 1 : 100, as well as rabbit polyclonal anti bodies against nestin (1 : 100; Chemicon, USA), beta IIItubulin (1 : 100, Sigma, USA) and GFAP (1 : 100, Abcam, USA). The cells were incubated with the pri mary antibodies for at least 12 hours (overnight) at 4°C. As far as secondary antibodies goat antirabbit antibody conjugated to CY2 (1 : 300) and goat anti bodies to antimouse immunoglobulins labeled with CY3 (1 : 300) (Chemicon, USA) or Dylight 488 from the company Jackson Immunoresearch (USA) (1 : 400) were used. The nuclei were stained with DAPI (Merck, USA) in a concentration of 1 mg/mL. Coverslips were mounted in 2% propylgallate and visualized under Axiovert 200 M microscope (Carl Zeiss, Germany) equipped with Lieca DFC 420 C camera (Germany), as well as using a confocal scan ning microscope Leica TCS CL (Germany). For immunofluorescent analysis of surface antigen SSEA–4 the live cells were incubated with the mouse monoclonal antibodies against SSEA4 (Chemicon, USA) at a dilution of 1 : 50 for 30–40 min and fixed for 15 min with 4% formaldehyde solution (Sigma, USA) in PBS. Then, the cells were washed in PBS, contain ing 0.1% Tween20 and incubated with 1% solution of BSA in PBS to block a nonspecific binding of the anti bodies. The rest was performed according to standard protocol. Reverse TranscriptionPolymerase Chain Reaction (RTPCR) analysis. For genes expression analysis,
238
ZEMELKO et al.
total RNA was extracted from the cultured cells using RNAeasy Micro Kit (QIAGEN, USA) according to the manufacturer’s protocol. Synthesis of cDNA was performed from 500 ng of total RNA using RevertAid H Minus First Strand cDNA Synthesis Kit (Fermen tas, Lithuania) following manufacturer’s instructions. Polymerase chain reaction (PCR) amplification was carried out using Taq DNA polymerase (Fermentas, Lithuania) in the thermocycler Cyclo Temp (STM, Russia). Each reaction had the following cycling parameters: denaturation at 93°C for 45 seconds, annealing of primers at 56–70°C for 1 min, chain elongation at 72°C for 1.5 min, and then terminating chain elongation at 72°C for 10 min for 26–40 cycles. The cycles were preceded by a hot startdenaturation at 93°C for 3 min, annealing of primers at 56–70°C for 2 min, chain extension at 72°C for 1.5 min. For the genes expression analysis the following primers were used: NeuN direct 5'TTTA ACGAGCGGGGCTCCAAG3' and reverse 5'TTC ATGGTCCGAGAAGGAAACG3', annealing tem perature 65°C (amplification product size 578 bp); βIIItubulin direct 5'GGATTCGGTC CTGGATGTGG3' and reverse 5'AAGCCG GGCATGAAGAAGTG3', temperature annealing 61°C (471 bp); MAP2 direct 5'GGGAAG AGTGGTACCTCAACA3' and reverse 5'TTG GCTGTCAATCTTGACAT3', annealing tempera ture 61°C (525 bp); NFH direct 5'GAA AAGCACCAAGGACTCAC3' and reverse 5'GGTGACTTTGCCTCTTCCTT3', annealing tem perature 57°C (576 bp); GFAP direct 5'GTG GTACCGCTCCAAGTTTGCAG3' and reverse 5'AATGGTGATCCGGTTCTCCTC3', annealing temperature of 65°C (373 bp), neurotrophin3 (NT3) direct 5'TACGCGGAGCATAAGAGTCAC3' and reverse 5'GGCACACACACAGGACGTGTC3', annealing temperature of 65°C (333 bp), neurotro phin4 (NT4) direct 5'CTTTCG GGAGTCAGCAGGTGC3' and reverse 5'CAGG CAGTGTCAATTCGAATCC3', the annealing tem perature of 65°C (399 bp). As a quantitative control of RNA and a control of DNA contamination during reverse transcription (RT) gene αactin was used: forward primer 5'GCCGAGCGGGAAATCGTGCGT3' and reverse 5'CGGTGGACGATGGAGGGGCCG3', annealing temperature of 70°C (507 bp). Electrophoresis of the PCR amplified products was performed in 2% agarose gel containing ethidium bromide (Sigma, USA) and band sizes were compared with a100kilobase DNA ladder (Gibco, USA).Visualization was performed in UV light (wavelength 302 nm) using a device Transil luminator, recording images with a digital camera Canon. Identification of brainderived neurotrophic factor (BDNF) synthesis in the differentiated MSCs using ELISA. Immunoferment assay of a culture medium of the neuraldifferentiated MSCs was performed using a
DuoSet ELISA Development System for human BDNF (R & D systems, USA). According to the man ufacturer’s recommendations the precoated 96well plate, containing the culture medium from the cells before and after differentiation, was incubated for 2 h at room temperature. The wells with the standard recombinant human BDNF were used as a positive control. Then, a biotinylated antihuman BDNF antibody was added to the wells and incubated for 2 h at room temperature. To detect the bound antibody streptavidin conjugated to horseradish peroxidase was used. A mixture of tetramethylbenzidine with perox ide served as a reaction substrate. At every stage of pro cessing, the wells were washed 3 times with 0.05% of Tween20 in PBS. The color reaction was recorded by measuring the absorbance of the well’s solution at 450 nm wavelength by immunochemical analyzer FLUOROFOT (Germany). All the measurements were performed in duplicates. RESULTS AND DISCUSSION Replacement and regenerative cell therapy shows a great promise in the treatment of many diseases. The use of adult autologous MSCs is not associated with the ethical controversy, and also eliminates the risk of tumorigenicity (Murphy et al., 2008) and immune rejection that limit the clinical use of human ESCs. A potential use of the MSCs, as an alternative source of stem cells for the treatment of neurodegenerative dis eases, has investigated for a long time (Woodbury et al., 2002; Jori et al., 2005). To date, some effective protocols of neural differentiation of the MSCs in vitro described (Blondheim et al., 2006; Borlongan et al., 2010; Jang et al., 2010), including those based on the use of 5azacytidine (Kang et al., 2003; Lee and Yoon, 2008) and morphogen SHH together with growth factor FGF8 (Long et al., 2005; Trzaska et al., 2007). However, the present work is the first systematic analysis of the neurogenic potential of the three types of human MSCs. In a paper (Zhang et al., 2012) in similar design were compared BMSCs and ADCS, but not eMCSs. Identification and phenotypic analysis of the iso lated human MSCs. An immunophenotype of the MSCs has been analyzed in accordance with the rec ommendations of the International Society for Cellu lar Therapy (ISCT) (Dominici et al., 2006). To con firm that the isolated from BM, ADT and endometrium primary cultures are populations of the MSCs, we analyzed the content of theMSCs surface markers by flow cytometry. The table reflects the quantitative analysis of the expression of surface anti gens for the eMSCs and BMCSs and ADSCs. So, the derived eMSCs express surface markers such as the SD9, CD13, CD29, CD44, CD73, CD90, CD105, CD146, and do not express the hematopoietic markers CD11a, CD19, CD 34, CD45, CD117 and HLADR (class II), and the BMSCs and ADSCs express CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
NEUROGENIC POTENTIAL OF HUMAN MESENCHYMAL STEM CELLS
CD105, CD73 and CD90, and do not express CD45, CD34 and CD14. These results are consistent not only with the ISCT criteria for phenotyping of human MSCs based on the expression of their surface mark ers, but also with numerous data of other researchers (Pettinger et al., 1999; Buhring et al., 2007; Gargett and Masuda, 2010; Husein and Thiemermann, 2010). Morphologically, all derived MSC cultures formed a monolayer of fibroblastlike cells with the characteris tic circular twists. Immunofluorescent analysis of all obtained MSC cultures showed phenotypic expression (about 70%) of multipotency marker SSEA4 (Fig. 1, BM a, ADT a and eMSC a), a marker of early neuronal progenitors nestin (Fig. 1, BM b, ADT b and eMSC b), as well as a marker of late neuronal precursors betaIIItubulin (Fig. 1, BM c, ADT c and eMSC c) Expression of anti gen SSEA4 is essential characteristic of ESCs. But in contrast to the uniformity of SSEA4 distribution across the membrane in the ESCs, localizations of this surface antigen in the MSCs are concentrated in areas of focal adhisions. The expression of this multipotency marker has been already shown for BMSCs (Gang et al., 2007), ADSCs (Riekstina et al., 2009) and eMSCs (Patel et al., 2008; Zemelko et al., 2011). In 2006, it was discovered that BMSCs express 12 neuronal genes (including the marker of early of neuronal progenitors nestin), 8dopaminergic genes, and 11 neuronal transcription factors before the cells go to neurogenesis (Blondheim et al., 2006). Later, the expression of nestin in placenta MSCs (Chen et al., 2008), immunofluorescent staining of ectoderm marker betaIIItubulin in umbilical cord blood MSCs (Zwart et al., 2008) and the expression of both these markers in eMSCs (Zemelko et al., 2011) were found. In this paper, for the first time, by immunoflu orescent and RTPCR analysis (Fig. 3a–3c) was shown that the BMSCs, the ADSCs and the eMSCs cultures all express not only the marker of early neu ronal precursors nestin, but also the marker of late neuronal precursors betaIIItubulin. Thus, in our opinion, it is not competent to consider these antigens as markers of neural induction in MSCs. In the ADSC and eMSC cultures (but not the BMSC), we have also found that some single cells express antigen of neuronal intermediate filaments (NFH) (Fig. 1, ADT d and eMSC d, respectively). There is a research where the authors (Blondheim et al., 2006) showed low levels of expression of NFH in undifferentiated BMSCs, although in this study, we have not confirmed it. Characteristic phenotypic expression of the antigens SSEA4, nestin and beta IIItubulin in the cytoskeleton of the derived MSCs indicates their multipotent plasticity and neural pre disposition. To induce neural differentiation of the obtained MSC cultures agents such as 5azacytidine (a chemi cal compound that causes DNA demethylation), bFGF, EGF, SHH, FGF8, RA and IBMX were used CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
239
Quantitative analysis of the expression of surface antigens for the derived eMSCs, BMSCs and ADSCs Marker’s distribution, %
Surface marker
BMSC
ADSC
eMSC
CD9 CD13 C29 CB44 CD73 CD90 CD105 CD146 CB11a CD19 CD34 CD45 CD117 HLADR
N.d. N.d. N.d. N.d. 66.64 ± 19.5 72.64 ± 17.2 59.21 ± 17 20.96 ± 14.04 N.d. 4.03 ± 0.1 2.35 ± 0.1 0.07 ± 0.1 N.d. N.d.
N.d. N.d. N.d. N.d. 34.27 ± 19.42 57.94 ± 22.66 96.22 ± 3.59 21.33 ± 8.81 N.d. N.d. 0.55 ± 0.1 0.47 ± 0.1 N.d. N.d.
98.0 ± 1.2 99.75 ± 0.15 99.9 ± 0.1 99.6 ± 0.1 99.9 ± 0.1 95.35 ± 1.35 99.8 ± 0.1 94.45 ± 0.45 3.32 ± 2.76 0.28 ± 0.1 0.22 ± 0.2 0.8 ± 0.1 3.73 ± 2.91 0.93 ± 0.47
Note: N.d. not done.
in two experimental protocols (see “Materials and Methods”). Applying different protocols within the same type of MSC cells as well as for the MSCs from the different origins, we evaluated the efficiency of neural induction based on the transcription of marker genes and upregulation of marker proteins, such as MAP2, GFAP, NFH and NeuN. In addition, we ana lyzed the secretion by the neuraldifferentiated MSCs neurotrophic factors such as BDNF (by ELISA) and neurotrophin3 and neurotrophin4 (by RTPCR). Matrigel and laminin were used as the cells substrates for the differentiation of the MSCs. The stimulation of neural differentiation in all types of the MSCs led to a gradual change in the mor phology of the cells. Some cell’s body becomes more compact with the appearance of long thin processes. Change in morphology from fibroblastlike to neural for the BMSCs, ADSCs and eMSCs reflected in Figs. 2a–2c. Our results show that the most efficient neural induction in the BMSCs takes place without RA with prior treatment with 5azacytidine (protocol 1) (Fig. 3a). Protocol 2 is quite effective as well, where an important differentiation agent is mor phogen SHH, whose gradient plays an important role in the development and organization of the fetal brain of vertebrates (Johnson et al., 1994; Litingtung and Chiang, 2000; Kolpak et al., 2005). The summary expression of marker proteins MAP2, GFAP, NFH and NeuN in the differentiated BMSCs is reflected in Fig. 2, BM d–g. Unlike the BMSCs, for the ADSC and eMSC cells, which characterized by a high proliferative activity,
240
ZEMELKO et al. BM (a)
(b)
SSEA4
NESTIN
(c)
βIIITUBULIN
ADT (a)
(b)
(c)
(d)
SSEA4
NESTIN
βIIITUBULIN
NFH
(a)
(b)
(c)
(d)
SSEA4
NESTIN
βIIITUBULIN
NFH
eMSC
Fig. 1. Expression of phenotypic markers of the MSCs from bone marrow (BM), subcutaneous adipose tissue (ADT) and endometrium (eMSC). a–c—Markers SSEA4, nestin and betaIIItubulin respectively, d—expression of the antigen of neuronal intermediate fila ments (NFH). Scale bar: BM—25 (a, c) and 15 (b) µm, ADT—25 (a, c, d) and 15 (b) µm, eMSC—25 (a, c) and 15 (b, d) µm.
(Parker and Katz, 2006; Meng et al., 2007) RA, espe cially in combination with 5azacytidine, may be optional (see Fig. 3b) and (Fig. 2, ADT d–g) and for the eMSCs is a necessary agent for the neural induc tion (Fig. 2, eMSC d–i). It is known that RA influ ences the development of the central nervous system and regulates neural differentiation in the ESCs (Bain et al., 1995; Chen et al., 2010). Its action is based on the increased transcription of many neuronspecific genes with RAresponsive element in their promoters. It is conceivable that RA together with IBMX, that causes an increase in intracellular cAMP levels (Wang et al., 2007), play a key role in stopping the prolifera tion and stimulating the neural differentiation of the eMSCs. PTPCR analysis of analyzed genes and neu rotrophic factors revealed the basal expression of MAP2 and NT3 in the undifferentiated BMSCs and ADSCs. The induction of neural differentiation in these cells caused an increase of the expression level of MAP2, but almost did not change the expression level of NT3 (Figs. 3a, 3b). In this case there was no tran scription of NT4. Unlike the BMSCs and the ADSCs the neural induction in the eMSC culture stimulated the expression of NT4, especially on the laminin as a substrate (Fig. 3c).
ELISA assay of the culture medium of the neural differentiated MSCs to identify their BDNF synthesis revealed that the cells of different types have different basal levels of this factor. In the BMSC culture BDNF synthesis was not detected, the ADSCs express a small amount of BDNF, that in agreement with other data (Zhang et al., 2012), and the eMSCs express relatively high number of BDNF in control. Since the undiffer entiated MSCs from BM and ADT do not secrete (or slightly secrete) BDNF, to study the induction of BDNF in those cells were analyzed the culture medium of the differentiated BMSCs and ADSCs pre treated with 5azacytidine. The eMSCs, in our view, did not require the demethylation prior the differenti ation. The results of the induction in the secretion of BDNF in a process of directed neural differentiation showed that all three types of the MSCs have the potential to increase the secretion of this neurotrophic factor (Fig. 4). ELISA analysis of the impact of RA as an inducer of the differentiation led to the conclusion (Fig. 4) that the presence of RA is essential for the neural induction in the eMSCs, whereas for the differ entiation of the BMSCs and the ADSCSs its presence is not necessary. Moreover, for the enhancement of the secretion of BDNF in the eMCSs use of laminin as a substrate was crucial. CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
NEUROGENIC POTENTIAL OF HUMAN MESENCHYMAL STEM CELLS (a)
(b)
241
(c)
BM βIIITUBULIN (d)
(e)
(f)
(g)
MAP2
GFAP
NFH
NeuN
(a)
(b)
(c)
ADT βIIITUBULIN (d)
(e)
(f)
(g)
MAP2
GFAP
NFH
NeuN
(a)
(b)
(c)
eMSC βIIITUBULIN (d)
(e)
(f)
(g)
MAP2
GFAP DAPI
NFH
NeuN
(h)
(i)
βIIITUBULIN MAP2
GFAP MAP2
Fig. 2. Morphological changes (a, b, c) and expression of markers in the neuraldifferentiated MSCs from BM, ADT and eMSC (d, e, f, g, h, i). a—Undifferentiated cells, b—differentiated cells, c—betaIIItubulin staining, d—expression of microtubules associated pro tein MAP2, e—expression of glial fibrillary acidic protein (GFAP), for the eMSCs double staining GFAP (green) and DAPI (blue), f—expression of NFH, g—expression of neuronal nuclei protein (NeuN), h—double staininig of the betaIIItubulin (green) and MAP2 (red), i—double staining of GFAP (green) and betaIIItubulin (red). Phase contrast 10× (a, b). Scale bar: BM—15 (c), 25 (e, f) and 47.5 (d, g) µm, ADT—15 (c, d), 25 (f) and 47.5 (e, g) µm, eMSC—25 µm. Protocols 1 and 2 for the BMSCs and eMSCs and protocol 1 for the ADSCs were used. (see “Materials and Methods”).
CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
242
ZEMELKO et al. (a) (b) βIIITUBULIN MAP2 NFH NEUN NT3 NT4 βACTIN
(c)
R A
C on tro 5 l az an o 5 R az A aw ith R A
C on tro no l 5 5a az za 5 a no az a w RA ith R A
βIIITUBULIN NT4 βACTIN C R on A tro no l M 5 at a rig za R e o A l n la no m 5 in az in a e on
GFAP βIIITUBULIN MAP2 NFH NEUN NT3 NT4 βACTIN
Fig. 3. mRNA expression of neural genes in the control and the differentiated BMSCs (a), ADSCs (b) and eMSCs (c). Protocol 1 was followed.
Currently, there is an active research of RA role in the neural differentiation of stem cells, including MSCs (Kim et al., 2002; Scintu et al., 2006), and the comparative analysis of the influence of this factor on the neural differentiation of three types of the MSCs present a considerable interest. Components of the extracellular matrix including laminin, are important factors of an induction of a stem cells differentiation, including a neural differentiation (Kozhuharova et al., 2010; Suri and Schmidt, 2010). However, in a recent paper (di Summa et al., 2012) it have been shown that laminin does not affect the secretion of BDNF in ADSCs. In our study different response of the BMSCs, the ADSCs and eMSCs to laminin also illus BDNF 2.0
1.5
1.0
0.5
0
Control
RA
No RA
Fig. 4. Secretion of brainderived neurotrophic factor (BDNF) in the differentiated BMSCs (black bars), ADSCs (gray bars) and eMSCs (hatched bars). Protocol 1 was used. Vertical axis: relative amount of BDNF, pg/1 µg RNA.
trates that the neurogenic potential of these three types of the MSCs is different. Summarizing, we can say that initially all three MSCs cultures have the multipotent plasticity and predisposition to neurogenesis, based on the expres sion of markers of pluripotency SSEA4 and neuronal precursors (nestin and betaIIItubulin). Further analysis revealed that in the undifferentiated BMSCs and ADSCs there is the transcription of the neuronal marker genes MAP2 and NT3, and the eMSC culture have high basal level of synthesis of BDNF. Stimula tion of neural induction with the agents such as 5aza cytidine, bFGF, EGF, SHH, FGF8, RA and IBMX, including components of the extracellular matrix Matrigel and laminin showed the additional differ ences in neurogenic potential of the cells. Thus, the most effective BMSC neural differentiation took place without the participation of RA with 5azacytidine pretreatment, but for the eMSC cells RA was the nec essary agent of neural induction by stimulating the transcription of NF4 and increasing the secretion of BDNF. In case of the eMSCs the use of laminin as substrate was necessary, and incubation the cells with 5azacytidine was optional. As far as the ADSCs RA in combination with 5azacytidine induces activation of the expression of MAP2, which is consistent with the literature (Lopa tina et al., 2008), but decreased the secretion of BDNF. Thus, the effect of RA on the neural differen tiation of the ADSCs is ambiguous and requires fur ther research along with the study of its signaling path ways in the MSCs. The most common hypothesis of the therapeutic effect of transplanted MSCs based on a paracrine mechanism whereby transplanted cells release factors that promote a repair of damaged tissue. BDNF is one of the most active neurotrophins, which supports the viability of existing neurons and promotes the growth and differentiation of new neurons and synapses (Acheson et al., 1995; Zigova et al., 1998; Huang et al., CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
NEUROGENIC POTENTIAL OF HUMAN MESENCHYMAL STEM CELLS
2001). The high basal level of BDNF synthesis in the cultured eMSCs from menstrual blood, along with their high proliferative activity, neuronal predisposi tion and a noninvasive method of isolation are the powerful arguments for the use of the intact eMSCS as a substrate for a cell therapy to repair a nerve tissue. ACKNOWLEDGMENTS This work was supported by the Ministry of Health and Social Development of Russian Federation (State contract no. K32NIR/1113) under the program “Stem Cells”. REFERENCES Acheson, A., Conover, J.C., Fandl, J.P., DeChiara, T.M., Russell, M., Thadani, A., Squinto, S.P., Yancopoulos, G.D., and Lindsay, R.M., A BDNF autocrine loop in adult sen sory neurons prevents cell death, Nature, 1995, vol. 374, pp. 450–453. Anisimov, S.V., Cell therapy of Parkinson’s disease: IV. Risks and prospects, Usp. Gerontol., 2009, vol. 22, no. 3, pp. 418–439. Bain, G., Kitchens, D., Yao, M., Huettner, J.E., and Gott lieb, D.I., Embryonic stem cells express neuronal properties in vitro, Dev. Biol., 1995, vol. 168, pp. 342–357. Blondheim, N.R., Levy, Y.S., BenZur, T., Burshtein, A., Cherlow, T., Kan, I., Barzilai, R., BahatStromza, M., Bar hum, Y., Bulvik, S., Melamed, E., and Offen, D., Human mesenchymal stem cells express neural genes, suggesting a neural predisposition, Stem Cells Dev., 2006, vol. 15, pp. 141–164. Borlongan, C.V., Kaneko, Y., Maki, M., Yu, S.J., Ali, M., Allickson, J.G., Sanberg, C.D., KuzminNichols, N., and Sanberg, P.R., Menstrual blood cells display stem celllike phenotypic markers and exert neuroprotection following transplantation in experimental stroke, Stem Cells Dev., 2010, vol. 19, pp. 439–452. Buhring, H.J. and Battula, V.L., Novel markers for the pro spective isolation of human MSC, Ann. N.Y. Acad. Sci., 2007, vol. 1106, pp. 262–271. Chen, C.W., Liu, C.S., Chiu, I.M., Shen, S.C., Pan, H.C., Lee, K.H., Lin, S.Z., and Su, H.L., The Signals of FGFs on the neurogenesis of embryonic stem cells, J. Biom. Sci., 2010, vol. 17, p. 33. Chen, I., He, D.M., and Zhang, Y., The Differentiation of human placentaderived mesenchymal stem cells into dopaminergic cells in vitro, Cell Mol. Diol. Lett., 2009, vol. 14, pp. 528–536. da Silva Meirelles, L., Chagastelles, P.C., and Nardi, N.B., Mesenchymal stem cells reside in virtually all postnatal organs and tissues, J. Cell Sci., 2006, vol. 119, pp. 2204– 2213. di Summa, P.G., Kalbermatten, D.F., Raffoul, W., Terenghi, G., and Kingham, P.J., Extracellular matrix mol ecules enhance the neurotrophic effect of Schwann celllike differentiated adiposederived stem cells and increase cell survival under stress conditions, Tissue Eng. Part. A, 2013, vol. 19, pp. 368–379. CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
243
Dmitrieva, R.I., Minullina, I.R., Bilibina, A.A., Tarasova, O.V., Anisimov, S.V., and Zaritskey, A.Y., Bone marrow and subcutaneous adipose tissuederived mesen chymal stem cells: differences and similarities, Cell Cycle, 2012, vol. 11, pp. 377–383. Dominici, M., Le, Blanc, K., Mueller, I., SlaperCorten bach, I., Marini, F., Krause, D., et al., Minimal criteria for defining multipotent mesenchymal stromal cells. the inter national society for cellular therapy position statement, Cytotherapy, 2006, vol. 8, pp. 315–317. Gang, E.J., Bosnakovski, D., Figueiredo, C.A., Visser, J.W., and Perlingeiro, R.C., SSEA4 identifies mes enchymal stem cells from bone marrow, Blood, 2007, vol. 109, pp. 1743–1751. Gargett, C.E. and Masuda, H., Adult stem cells in the endometrium, Mol. Hum. Reprod., 2010, vol. 16, pp. 818– 834. Huang, E.J. and Reichardt, L.F., Neurotrophins: roles in neuronal development and function, Annu. Rev. Neurosci., 2001, vol. 24, pp. 677–736. Husein, K.S. and Thiemermann, C., Mesenchymal stromal cells: current understanding and clinical status, Stem Cells, 2010, vol. 28, pp. 585–596. Jang, S., Cho, H.H., Cho, Y.B., Park, J.S., and Jeong, H.S., Functional neural differentiation of human adipose tissue derived stem cells using bFGF and forskolin, BMC Cell Biol., 2010, vol. 16, pp. 11–25. Johnson, R.L., Riddle, R.D., Laufer, E., and Tabin, C., Sonic hedgehog: a key mediator of anteriorposterior pat terning of the limb and dorsoventral patterning of axial embryonic structures, Biochem. Soc. Trans., 1994, vol. 22, pp. 569–574. Jori, F.P., Napolitano, M.A., Melone, M.A., Cipollaro, M., Cascino, A., Altucci, L., Peluso, G., Giordano, A., and Galderisi, U., Molecular pathways involved in neural in vitro differentiation of marrow stromal stem cells, J. Cell Biochem., 2005, vol. 94, pp. 645–655. Kang, S.K., Lee, D.H., Bae, Y.C., Kim, H.K., Baik, S.Y., and Jung, J.S., Improvement of neurological deficits by intracerebral transplantation of human adipose tissue derived stromal cells after cerebral ischemia in rats, Exp. Neurol., 2003, vol. 183, pp. 355–366. Kim, Â., Seo, J.H., Bubien, J.K., and Oh, Y.S., Differenti ation of adult bone marrow stem cells into neuroprogenitor cells in vitro, Neuroreport, 2002, vol. 13, pp. 1185–1188. Kolpak, A., Zhang, J., and Bao, Z.Z., Sonic hedgehog has a dual effect on the growth of retinal ganglion axons depending on its concentration, J. Neurosci., 2005, vol. 25, pp. 3432–3441. Kozhukharova, I.V., Fridlyanskaya, I.I., Zemelko, V.I., Kovaleva, Z.V., Pugovkina, N.A., Alekseenko, L.L., Kharchenko, M.V., Aksenov, N.D., Shatrova, A.N.., Grin chuk, T.M., Anisimov, S.V., and Nikolsky, N.N., Genera tion of dopamine neurons from human embryonic stem cells in vitro, Cell Tiss. Biol., 2010, vol. 4, no. 5, pp. 411– 418. Lee, T.H. and Yoon, J.G., Intracerebral transplantation of human adipose tissue stromal cells after middle cerebral artery occlusion in rats, J. Clin. Neurosci., 2008, vol. 15, no. 8, pp. 907–912.
244
ZEMELKO et al.
Litingtung, Y. and Chiang, C., Control of SHH activity and signaling in the neural tube, Dev. Dyn., 2000, vol. 219, pp. 143–154. Long, X., Olszewski, M., Huang, W., and Kletzel, M., Neu ral cell differentiation in vitro from adult human bone mar row mesenchymal stem cells, Stem Cells Dev., 2005, vol. 14, pp. 65–69. Lopatina, T.V., Kalinina, N.I., Revischin, A.V., Beme, A.A., Spirova, I.A., Pavlova, G.V., and Parfenova, E.V., Induction of neural differentiation of stro mal cells of adipose tissue, Klet. Transplantol. Tkan. Inzhener., 2008, vol. 3, no. 4, pp. 50–55. Meng, X., Ichim, T.E., Zhong, J., Rogers, A., Yin, Z., Jack son, J., Wang, H., Ge, W., Bogin, V., Chan, K.W., Thébaud, B., and Riordan, N.H., Endometrial regenerative cells: a novel stem cell population, J. Transl. Med., 2007, vol. 5, pp. 57–66. Molero, A.E., Gokhan, S., Gonzalez, S., Feig, J.L., Alex andre, L.C., and Mehler, M.F., Impairment of develop mental stem cellmediated striatal neurogenesis and pluri potency genes in a knockin model of Huntington’s disease, Proc. Natl. Acad. Sci. USA, 2009, vol. 106, pp. 21900– 21905. Murphy, M.P., Wang, H., Patel, A.N.., Kambhampati, S., Angle, N., Chan, K., Marleau, A.M., Pyszniak, A., Carrier, E., Ichim, T.E., et al., Allogeneic endometrial regenerative cells: an ‘off the shelf solution’ for critical limb ischemia?, J. Transl. Med., 2008, vol. 6, pp. 45–52. Parker, A.M. and Katz, A.J., Adiposederived stem cells for the regeneration of damaged tissues, Expert. Opin. Biol. Ther., 2006, vol. 6, pp. 567–578. Patel, A.N., Park, E., Kuzman, M., Benetti, F., Silva, F.J., and Allickson, J.G., Multipotent menstrual blood stromal stem cells: isolation, characterization, and differentiation, Cell Transplant., 2008, vol. 17, pp. 303–311. Pittenger, M.F., Mackay, A.M., Beck, S.C., Jaiswal, R.K., Douglas, R., Mosca, J.D., Moorman, M.A., Simonetti, D.W., Craig, S., and Marshak, D.R., Multilin eage potential of adult human mesenchymal stem cells, Sci ence, 1999, vol. 284, pp. 143–147. Prockop, D., Repair of tissues by adult stem/progenitor cells (MSCs): controversies, myths, and changing para digms, Mol. Ther., 2009, vol. 17, pp. 939–946. Riekstina, U., Cakstina, I., Parfejevs, V., Hoogduijn, M., Jankovskis, G., Muiznieks, I., Muceniece, R., and Ancans, J., Embryonic stem cell marker expression pattern in human mesenchymal stem cells derived from bone mar row, adipose tissue, heart and dermis, Stem Cell Rev., 2009, vol. 5, pp. 378–386.
Schraufstatter, I.U., Discipio, R.G., and Khaldoyanidi, S., Mesenchymal stem cells and their microenvironment, Front. Biosci., 2011, vol. 17, pp. 2271–2288. Scintu, F., Reali, C., Pillai, R., Badiali, M., Sanna, M.A., Argiolu, F., Ristaldi, M.S., and Sogos, V., Differentiation of human bone marrow stem cells into cells with a neural phe notype: diverse effects of two specific treatments, BMC Neurosci., 2006, vol. 16, pp. 7–14. Suri, S. and Schmidt, C.E., Cellladen hydrogel constructs of hyaluronic acid, collagen, and laminin for neural tissue engineering, Tissue Eng. Part A, 2010, vol. 16, pp. 1703– 1716. Thornell, L.E., Lindstöm, M., Renault, V., Klein, A., Mouly, V., Ansved, T., ButlerBrowne, G., and Furling, D., Satellite cell dysfunction contributes to the progressive muscle atrophy in myotonic dystrophy type 1, Neuropathol. Appl. Neurobiol., 2009, vol. 35, pp. 603–613. Trzaska, K.A., Kuzhikandathil, E.V., and Ramashwar, P., Specification of a dopaminergic phenotype from adult human mesenchymal stem cells, Stem Cells, 2007, vol. 25, pp. 2797–2808. Wang, T.T., Tio, M., Lee, W., Beerheide, W., and Udolph, G., Neural differentiation of mesenchymallike stem cells from cord blood is mediated by PKA, Biochem. Biophys. Res. Commun., 2007, vol. 357, pp. 1021–1027. Woodbury, D., Reynolds, A., and Black, I.B., Adult bone marrow stromal stem cells express germline, ectodermal, endodermal and mesodermal genes prior to neurogenesis, J. Neurosci. Res., 2002, vol. 96, pp. 908–917. Zemelko, V.I., Grinchuk, T.M., Domnina, A.P., Artsyba sheva, I.V., Zenin, V.V., Kirsanov, A.A., Bichevaya, N.K., Korsak, V.S., and Nikolsky, N.N., Multipotent mesenchy mal stem cells of desquamated endometrium: isolation, characterization, and application as a feeder layer for main tenance of human embryonic stem cells, Cell Tiss. Biol., 2011, vol. 53, no. 12, pp. 919–929. Zhang, H.T., Liu, Z.L., Yao, X.Q., Yang, Z.J., and Xu, R.X., Neural differentiation ability of mesenchymal stromal cells from bone marrow and adipose tissue: a com parative study, Cytotherapy, 2012, vol. 14, pp. 1203–1214. Zigova, T., Pencea, V., Wiegand, S.J., and Luskin, M.B., Intraventricular administration of BDNF increases the number of newly generated neurons in the adult olfactory bulb, Mol. Cell. Neurosci., 1998, vol. 11, pp. 234–245. Zwart, I., Hill, A.J., Girdlestone, J., Manca, M.F., Navar rete, R., Navarrete, C., and Jen, L.S., Analysis of neural potential of human umbilical cord bloodderived multipo tent mesenchymal stem cells in response to a range of neu rogenic stimuli, J. Neurosci Res., 2008, vol. 86, pp. 1902– 1915.
CELL AND TISSUE BIOLOGY
Vol. 7
No. 3
2013
