Stem Cell Rev and Rep (2010) 6:512–522 DOI 10.1007/s12015-010-9177-7

Reprogramming of Human Umbilical Cord Stromal Mesenchymal Stem Cells for Myogenic Differentiation and Muscle Repair Çetin Kocaefe & Deniz Balcı & Burcu Balcı Hayta & Alp Can

Published online: 28 July 2010 # Springer Science+Business Media, LLC 2010

Abstract Human umbilical cord stromal mesenchymal stem cells (hUCS-MSCs) have the potential to differentiate into numerous cell types including epithelial cells, neurons and hepatocytes in vitro, in addition to mesenchyme-derived cells such as osteocytes, chondrocytes and adipocytes. One important property of these cells is the lack of type II major histocompatibility complex class molecules, thus allowing them to be considered as an excellent candidate for transplantations. Besides the use of 5-azacytidine as a supraphysiological inducer of myogenic transformation, no study has been published to date addressing the myogenic transformation efficiency of hUCS-MSCs by using a gene transfection strategy and/or co-culture with muscle cell lines. Here, we demonstrate the reprogramming efficiency of these cells, which differentiate into myocytes in vitro by MyoD transcription factor, the master regulator of skeletal muscle

differentiation. Once induced via MyoD expression, hUCSMSCs exhibited many cellular signs of myogenic conversion within 5 days and became capable of forming multinucleated myofibers, which exhibited all functional markers of fusion machinery such as β-catenin, neural cell adhesion molecule and M-cadherin as well as muscle cell-specific structural proteins including desmin, α-actinin, dystrophin, myosin heavy chain, and myoglobin together with muscle-specific enzyme, creatinine phosphokinase. Furthermore, programmed hUCS-MSCs were also capable of fusing with rat primary myoblasts to form heterokaryonic myotubes. Taken together, this study demonstrates the success of a novel cell reprogramming approach to be further evaluated at the in vivo level for use in restoring the defective dystrophin function as intrinsically found in the skeletal muscle fibers of Duchenne muscular dystrophy patients.

The authors declare no potential conflicts of interest.

Keywords Umbilical cord MSCs . Myocyte . MyoD . Myogenic differentiation . Myotube

Ç. Kocaefe : B. Balcı Hayta Department of Medical Biology, Hacettepe University School of Medicine, Ankara, Turkey Ç. Kocaefe e-mail: [email protected] B. Balcı Hayta e-mail: [email protected] D. Balcı Ankara University Biotechnology Institute, Ankara, Turkey e-mail: [email protected] A. Can (*) Department of Histology and Embryology, Ankara University School of Medicine, Ankara University Stem Cell Institute, Sihhiye, 06100 Ankara, Turkey e-mail: [email protected]

Introduction Over the last years, human umbilical cord stromal mesenchymal stem cells (hUCS-MSCs) have been extensively investigated to understand their stem cell properties [1]. They gained increasing interest with regard to their potential beneficial use for therapeutical applications in regenerative medicine. hUCS-MSCs were shown to display similar or higher degree of plasticity compared to bone marrow-derived MSCs in terms of differentiating into certain lineages in vitro [2, 3], and their limited proliferative capacity can be regarded as a biosafety advantage. hUCSMSCs can be readily isolated in large amounts from the umbilical cord during delivery without a major need for in

Stem Cell Rev and Rep (2010) 6:512–522

vitro amplification. Their lack of expression of the major histocompatibility complex (MHC) class I and II antigens [1, 4] render them to be highly tolerable cells in transplantations. The expression of certain embryonic stem cell markers on their naïve forms and their sharing of numerous cellular markers with MSCs bring evidence for their multipotency, which has eventually led many groups to evaluate their potency in experimental cellular therapies [5–12]. Genetically inherited myopathies and especially Duchenne muscular dystrophy (DMD) represent the most common genetic disorders. From the gene replacement therapy perspective, muscle tissue already harbors its own obstacles. One important approach to restore the expression of the missing gene in inherited myopathies emerges from the developmental nature of the muscle itself. The mature muscle fiber is a syncytium of committed myoblasts, fused to build a multinucleated myotube and thus, a heterokaryon joining into a myofiber bringing in an intact gene is capable of restoring the defect within the syncytium [13, 14]. While this heterokaryon notion formulates the proof of the concept for the cell therapy approach, numerous cell sources have been tested to achieve successful replacement, starting very early from the myoblasts as the natural candidates, [15, 16] to bone marrowderived stromal cells [17, 18], and mesoangioblasts [19]. Examining the differentiation potential of stem cells into certain lineages still stands as one of the most common and standard ways to address stemness capacity. Initiating the transformation of stem cells to a certain cell type(s) is typically accomplished by culturing them in defined culture conditions, by which stem cells are faced with an extraordinary type of chemicals and organic substances that are capable of initiating the investigated differentiation program but normally do not exist in vivo, at least in suggested doses. Among those chemicals, 5-azacytidine (5-AZC), as a potent genomic DNA demethylating agent, is used to induce myocyte differentiation of hUCS-MSCs in particular for skeletal muscle [20] or cardiomyocyte differentiation [21–24]. The most probable mechanism of action is the supraphysiological and nonspecific induction of transcription throughout the genome, resulting in myogenic conversion that is due to the dominance of the myogenic program over the other differentiation cascades. Given the fact that hUCS-MSCs possess true myofibroblastic properties, in the present study, we designed a rather physiological approach for testing the in vitro myogenic transformation efficiency of hUCS-MSCs using MyoD gene transfer to examine whether they could be converted to myogenic differentiation through myotube formation, performing fusion and displaying musclespecific cell markers. Additionally, we cultured hUCSMSCs in direct association with rat primary myoblasts to examine to what extent hUCS-MSCs can be induced to differentiate into myoblastic lineage. Here, we show that hUCS-MSCs can be successfully reprogrammed as

513

evidenced by the cellular consequences of MyoD expression, a master regulator of skeletal muscle differentiation. hUCSMSCs expressed β-catenin, neural cell adhesion molecule (NCAM) and M-cadherin, which are involved in fusion machinery, and exhibited fused, elongated multinucleated myotubes expressing various markers of myogenic differentiation including dystrophin, desmin, α-actinin, myosin heavy chain (MyHC), muscle-specific creatine phosphokinase (CPK-M) and myoglobin (MB). In addition, reprogrammed hUCS-MSCs also contributed to the multinucleated myotubes formed by rat primary myoblasts.

Materials and Methods Isolation and Expansion of hUCS-MSCs Umbilical cords of full-term deliveries (n=18) were transferred to the laboratory under sterile conditions. Ethical approval was obtained from the institutional ethical review board (approval no. 69-1780-2005). hUCS-MSCs were isolated and expanded as described elsewhere [3]. Following the second passage, cells were used for transfection or coculture experiments. Prior to reprogramming, fluorescenceactivated cell sorting was applied to cell suspensions for the validation and further isolation of hUCS-MSCs using certain MSC markers such as CD105, CD44, and CD73 (eBioscience, Inc., San Diego, CA), all conjugated with either fluorescein isothiocyanate (FITC) or phycoerythrin (PE). Following transfection and induction of differentiation as described below, the growth medium was replaced with the differentiation medium, which consisted of a 1:1 mixture of DMEM/Ham’s Nutrient Mixture F12 (1:1) (Sigma) and 2% horse serum (Sigma) supplemented with 2 mM L-glutamine (Sigma) and 1% (w/v) penicillin and streptomycin (Sigma). Transfection and Co-Culture Experiments Three different sets of experiments were conducted to elucidate the myogenic differentiation potency of hUCS-MSCs in vitro. (i) Transfection of hUCS-MSCs alone using adenoviral MyoD expression vectors. First- generation serotype five adenovirus vectors were employed to express mouse MyoD cDNA driven by cytomegalovirus (CMV) promoter. The production methods have been described elsewhere [25]. Downstream to the MyoD cDNA, the expression cassette also included an IRES (Internal Ribosomal Entry Site) and an enhanced green fluorescent protein (EGFP) reporter to follow the transgene expression. Optimal transfection efficiency of the hUCS-MSCs was investigated in 24-well tissue culture dishes by transfecting 2×104 cells/well at

514

varying multiplicity of infection (MoI) values of adenoviral vectors. A GFP expressing serotype five recombinant adenovirus was used in the negative control experiments. In transfection studies, 2×104 cells were plated on poly L-lysine (Sigma)-coated cover slips in 24-well plates (∼60-70% subconfluence) and were transfected 24 h later with 50 MoI of adenoviral vectors to express MyoD. Twenty-four hours following transfection, the growth media were replaced with the differentiation media and the cells were harvested as follows: samples for morphological observations were fixed using 4% paraformaldehyde (PFA) solution (w/v) in phosphate buffered saline (PBS) in 12, 24, 48, 72 and 96-hour intervals following switching to the differentiation medium. Samples for protein expression experiments (western blots) were trypsinized at predetermined time points and collected in 1.5 mL centrifuge tubes and immediately sonicated in protein loading buffer and kept at −80°C. Samples for the RNA expression studies were harvested using 0.5 mL of Trisol solution (Invitrogen) at predetermined time points and the samples were kept at −80°C until the concerted extraction of the total RNA as described below. (ii) Co-culture of native hUCS-MSCs with C2C12 myoblasts. In C2C12 co-culture experiments, hUCSMSCs were plated at a density of 104 cells/well on poly L-lysine coated cover slips in 24-well tissue culture dishes and an equal amount of C2C12 mouse myoblasts [26]. By doing so, 100% confluency was achieved within 24 h incubation in proliferation medium. Upon confluency, the proliferation medium is replaced by the differentiation medium with and without the supplementation of 0.5, 1, 2 and 4 mM dibutyryl cyclic adenosine monophosphate (dbcAMP) (Sigma). Co-cultures were maintained constantly and the medium was refreshed in every 48 h. Following the induction of differentiation at days 1, 2, 3, 4 and 5, the co-cultures were fixed with 4% PFA solution. (iii) Co-culture of MyoD transfected hUCS-MSCs with rat primary myoblasts. The isolation of the primary rat myoblasts from newborn pups and all procedures were performed according to an institution-approved protocol and under strict biological containment. Briefly, lower extremity muscles were dissected meticulously under stereomicroscope and were subjected to collagenase type 1 (La Roche Ltd., Basel, Switzerland) and dispase (Sigma) digestion (0.5% w/v) at 37°C for 15 min. Mononuclear cell extracts were harvested following retrieval through a 30 μm mesh and Ficoll gradient and immediately plated on matrigel (BD Biosciences, NJ)-coated tissue culture dishes. Primary myoblasts were expanded in DMEM/Ham’s

Stem Cell Rev and Rep (2010) 6:512–522

Nutrient Mixture F12 medium (1:1) supplemented with 20% fetal calf serum (FCS) and 2% Ultroser G (Pall, NY) as a serum supplement, 2 mM L-glutamine and 1% of penicillin and streptomycin. Co-culture with hUCS-MSCs was implemented as follows: hUCSMSCs were plated at a density of 1×104 cells/well on matrigel-coated glass cover slips in 24-well tissue culture plates to achieve a 50% confluency. 24 h later they were transfected with 100 MoI of adenoviral vectors to express MyoD. Twenty-four hours following transfection, cells were washed three times with growth medium to ensure the clearing of the viral particles, and then previously isolated primary rat myoblasts (2×104 cells/well) were added on to the culture dishes containing transfected hUCS-MSCs. Twenty-four hours later, the co-cultures were switched to the differentiation medium. The co-culture experiments were terminated by 4% PFA fixation at 1, 2, 3, 4 and 5 days following the induction of differentiation.

Fixation, Immunocytochemistry and Microscopy Four percent PFA-fixed cells (30 min at room temperature) were stained with antibodies for detecting the efficiency of posttransfectional reprogramming and de novo synthesis of muscle-cell specific proteins. Mouse monoclonal antibodies against MyoD1 (Millipore Co., Billerica, MA) (1:50 in PBS, 90 min at 37°C), β-catenin (BD Biosciences) (1:100 in PBS, 120 min at 37°C), α-actinin (ICN Biomedicals, Irvine, CA) (1:100 in PBS, 90 min at 37°C), human dystrophin (Nterminus) (Millipore) (1:50 in PBS, 90 min at 37°C), merosin laminin α-2 chain specific to human and rabbit (Novocastra), NCAM (CD56) (clone 123C3.D5, Neomarkers Inc., Fremont, CA) (1:100 in PBS, 90 min at 37°C), and desmin (clone DEU-10, Sigma) (1:100 in PBS, 90 min at 37°C) were used. Cy3 goat anti-mouse IgG (Zymed Laboratories Inc., San Francisco) was used as a secondary antibody. FITC-phalloidin labeling (specific to F-actin; 20 μg/mL; 20 min at 37°C) was applied to observe F-actin filaments. For nuclear labeling, anti-human nuclei mouse monoclonal antibody (Chemicon, Temecula, CA) (1:200; 60 min at 37°C) was applied, followed by incubation with Cy3 goat antimouse IgG (Zymed Laboratories Inc., San Francisco). As a second nuclear stain, Sytox green (Invitrogen Corporation, Carlsbad, CA) was applied to certain cover slips (10 μM for 30 min at 37°C). All immunofluorescent antibody labeling, GFP signals and dyes were examined using a Carl Zeiss LSM 510 Meta confocal laser scanning microscope (Carl Zeiss, Jena, Germany) equipped with 488-nm argon ion and 543-nm and 633-nm helium neon lasers. LSM-510 software

Stem Cell Rev and Rep (2010) 6:512–522

(version 3.2) was used to obtain three-dimensional images, which were then reconstructed using consecutive optical sections of various thicknesses (0.25–0.50 μm). The detection parameters, such as laser intensity, amplifier offset and gain, and pinhole diameter were fixed and kept at the same values for all specimens. Immunoblotting Analyses Selected CAMs and markers known to play a role in the myogenic differentiation and fusion process were documented at the protein level by immunoblotting. The total protein concentrations of the cellular extracts from time-course differentiated hUCS-MSCs were determined by bicinchoninic acid (BCA) Protein Assay (Pierce, Rockford, IL), and 50 μg of total protein was loaded onto the 12% sodium dodecyl (lauryl) sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel. Following the transfer of proteins onto the nitrocellulose membrane (Bio-Rad, Hercules, CA), equal loading was verified by ponceau-S staining. The probing of the membrane was achieved using rabbit anti β-catenin (Santa Cruz Biotechnology Inc., Santa Cruz, CA), goat anti M-cadherin (Santa Cruz Biotechnology), antibodies and mouse monoclonal antibodies against NCAM (CD56) (clone 123C3.D5, Neomarkers Inch.) and desmin (clone DE-U-10, Sigma) using 1:200 dilutions. Generation of the signal was accomplished following appropriate horseradish-peroxidase-conjugated secondary antibodies (1:8000; Sigma) using chemiluminescence detection system (ECL Plus Western blotting detection kit) (Amersham Biosciences AB, Sweden). RNA Isolation and Quantitative rtPCR Analyses All samples were treated together to avoid any experimental bias resulting from handling and extraction. Total RNA from the cell cultures was harvested as described previously [27]. Quality control was verified by denaturing agarose gel electrophoresis and optical density measurements at 260/ 320. One microgram of total RNA was reverse-transcribed into cDNA using oligo(dT) primers using Improm II reverse transcriptase (Promega, Madison, WI). Equal amount of cDNA was used for the real-time amplification of the target genes using Jumpstart SYBR Green mix (Sigma-Aldrich) according to the manufacturer’s recommendations on a Rotorgene 6000 (Corbett Life Science, Australia) real-time quantitative polymerase chain reaction (PCR) instrument. PCR products were double checked on the agarose gel electrophoresis and melting curve analysis to reaffirm the absence of any non-specific products and primer dimers. The primer pairs and the reaction conditions were designed to include at least one intron to avoid misamplification of any contaminated DNA. The sequences of the primer pairs and the reaction conditions are available

515

upon request by e-mail. The geometric mean of the expressions of β-actin transcription factor II D (TFIID) and β-actin have been used to normalize the expression of the myogenic differentiation-related transcripts, as suggested by Vandesompele et al. [28]. The observation period for the expression of MB and the muscle-specific isoform of CPKM were extended up to 10 days, and equal amount of the mid-log phase PCR products were run on agarose gel electrophoresis to reveal semi-quantitative expression information. Alternatively spliced isoforms of MyHC and NCAM are expressed in cells according to the physiological state. The NCAM primers used in this study are designed to target the extracellular FN3 domain specific to skeletal muscle, whereas the MyHC primers span the 35th/36th exons, which are amenable to amplify all MyHC isoforms.

Results Observations of Live Cells The presence of the GFP within the expression cassette facilitated the live observation of transfected cells. A weak expression of GFP was detectable starting from the 12th hour following transfection. Flat, polygonal-shaped hUCS-MSCs (Fig. 1a) exhibited spindle shapes and developed polarities with tendril-like extensions protruding from both ends (Fig. 1b) predominantly found after day 2. hUCS-MSCs remained in their native polygonal morphology when transfected with GFP expression vector (Fig. 1d). However, starting from the 3rd day after MyoD transfection, they developed multinucleated fused giant cells (Fig. 1e). Once a contact was established, these polar extensions also facilitated the fusion and formation of multinucleated myotubes (Fig. 1c, e and f). Formation of the Fusion Machinery The fusion of the committed cells required the formation of adhesion-type interactions between adjacent membranes. The GFP expressing transfected cells displayed an elongated and polarized morphology (Fig. 2a) with association of the expression of the three known key markers of myogenic fusion: NCAM (Fig. 2b, c), β-catenin (Fig. 2d–f), and Mcadherin (not shown). NCAM, a key CAM that is known to play a cardinal role in myoblast differentiation, was found expressed in a membranous protein (Fig. 2b) as well as at cell-cell contact areas (Fig. 2c). β-catenin immunostaining on transfected hUCS-MSCs clearly demonstrated the homophilic interactions between the MyoD transfected cells starting from the 12th hour after switching to the differentiation medium as shown in Fig. 2d. These homophilic β-catenin interactions can clearly be observed at cell-cell junctions (Fig. 2e) and

516

Stem Cell Rev and Rep (2010) 6:512–522

Fig. 1 MyoD-GFP transfected live hUCS-MSCs at day 0 (a), day 3 (b) and day 5 (c) as shown by DIC microscopy (a–c) and live fluorescent microscopy (e–f). Control hUCS-MSCs transfected with

GFP expression vector at day 5 (d). Note the elongated cells at days 3 and 5 displaying many multinucleated cells (arrow in e). Scale bars: 100 μm

delimit the cytoplasmic interaction rims between the fusing cells during 72 h post-transfection (Fig. 2f). The untransfected (GFP negative) cells did not show any staining for the above mentioned proteins of the fusion machinery.

MyoD expression was not detectable in native HUCSMSCs but a robust expression was noted within the first 12 h (fold changes for MyoD expression were estimates) and stayed stable throughout the observation period (Fig. 5). The concomitant expressions of the muscle-specific transcripts βcatenin, desmin, MyHC, MB, and muscle-specific isoform of CPK-M were evaluated at the mRNA level (Fig. 5). The time-course expression of β-catenin and desmin showed strong correlation with the western blot results and the expression of the MyHC protein, which became detectable starting from the 24th hour following transfection (Figs. 4 and 5). The expression of MB and CPK-M were fairly detectable early in the course of conversion; thus, the observation time was prolonged up to 10 days and the results were not quantitated but presented on the agarose gel electrophoresis (Fig. 5). Weak expression bands of MB and CPK-M became detectable starting from the 5th day posttransfection. While the expression of MB exhibited a stable pattern, a clear induction was observed in CPK-M expression at the 10th day. As a result of myogenic conversion, a decrease in cell cycle progression was expected. Ki-67 expression was documented as a marker of cell cycle arrest and clearly exhibited a gradual decrease over the observation period (Fig. 5) even though these cells were already in a confluent state.

Formation of the Muscle-Specific Structural Proteins The MyoD-induced myogenic commitment of hUCS-MSCs was also documented by the observation of the key skeletal muscle-specific structural proteins and differentiation markers. The expression of the major cytoskeletal muscle intermediate filament desmin was noticed starting from the 3rd day following transfection (Fig. 3a) as well as the synthesis of the extracellular matrix protein laminin (Fig. 3b) and α-actinin, the key structural component of the sarcolemma (Fig. 3c). The expression of the dystrophin protein was further analyzed using consecutive confocal images to demonstrate its typical subsarcolemmal localization in converting hUCS-MSCs (Fig. 3d–f). Dystrophin expression was detectable in fusing myotubular hUCS-MSCs starting from the 3rd day following MyoD transfection. Quantitative Expression Assays The time-course expressions of the three CAMs (β-catenin, M-cadherin and NCAM) involved in fusion and terminal differentiation as well as of desmin were also elucidated at the protein level by the immunoblotting technique (Fig. 4). The expression of β-catenin, which was firstly detected at the 12th hour, preceded the others. M-cadherin and NCAM expressions became prominent following the 3rd day post-transfection.

Co-Culture Experiments Observations on the native (non-transfected) hUCS-MSCs co-cultured with C2C12 myoblast cell line were prolonged up to 5 days. While the withdrawal of fetal serum in differentiation media initiated the fusion of the C2C12

Stem Cell Rev and Rep (2010) 6:512–522

517

Fig. 2 Evidences of functional fusion machinery formation. hUCSMSCs transform into long slender myotubes after MyoD transfection (a). They express NCAMs (red signals in b and c), where NCAMenriched sites coincide with cell-to-cell fusion sites (arrowheads in c). β-catenin containing cell adhesion sites (red signals in d, e and f) are

preferentially located in cell-to-cell contacts at an early time point (12 h) where cytoplasmic protrusions make several links with each other (arrowheads in d and e) or build broad appositions during fusion (72 h) (f). Blue signals correspond to cell nuclei at b and d. Scale bars: 50 μm

cells, surprisingly, they also partially continued to proliferate and the culture became overconfluent, not allowing clear observations following the 5th day. Upon the withdrawal of fetal serum factors, hUCS-MSCs ceased to proliferate and acquired elongated fusiform phenotypes. While the hUCS-MSCs maintained neat cell-to-cell contacts with C2C12 myoblasts (Fig. 6a arrowheads), they did not clearly contribute to multinucleated myotube structures (Fig. 6, upper row). Supplementation of the differentiation medium with 0.5, 1, 2 and 4 mM of dbcAMP did not exhibit any remarkable impact on myotube formation or C2C12 proliferation. Following MyoD transfection, hUCSMSCs clearly demonstrated fusion and heterokaryonic myotube formation with rat primary myoblasts. At least one myotube out of ten was observed to include human nuclei (Fig. 6, lower row).

components. However, the presence of an extraordinary number of 10-nm-thick intracytoplasmic filaments and gap junction type intercellular communications, as commonly observed at the interface of long cellular processes, gave credence to their classification as “unusual” smooth muscle cells having some kind of contractile properties. Studies demonstrated the expression of some muscle-specific cytoskeletal filaments in these cells [3], a finding that supports the notion that, rather than fibroblasts or smooth muscle cells, they are true myofibroblasts. Specifically, contractile proteins such as actin, nonmuscle myosin, desmin, and α-smooth muscle actin (α-SMA), a marker for myofibroblasts [31], are differentially expressed in hUCS-MSCs [3, 29, 32, 33], whereas muscle-myosin is lacking [29]. Based on the above findings, it was suggested that these cells possibly display specific functions related to both fiber synthesis and organized cell communication and contraction. The coexistence of vimentin and desmin in these cells [3, 29, 32] supports the hypothesis that they are intrinsically myofibroblasts. We have previously shown that WJ as a whole is composed of desmin and α-SMApositive, myofibroblast-like stromal cells [3]. Mitchell et al. [34] showed that SMA is expressed at similar levels in WJ cells grown on poly-D-lysine (PDL)/laminin matrix and in

Discussion Typical organization of organelles [29, 30] and the collagen-producing enzyme synthesis clearly demonstrated that hUCS-MSCs could be primarily regarded as fibroblasts responsible for the synthesis of collagen and other matrix

518

Stem Cell Rev and Rep (2010) 6:512–522

Fig. 3 MyoD-GFP transfected hUCS-MSCs exhibit certain structural proteins, all of which indicate that differentiating cells possess myogenic properties such as muscle-specific extracellular matrix protein laminin (red fibers in a), muscle-specific intermediate filament desmin (red signal in b), and structural sarcoplasmic protein α-actinin (red patches in c). They also express dystrophin as they transform into

myotubes detected as fine punctate staining beneath the sarcolemma (d: a single confocal section taken from mid nuclear plane). In e, successive confocal sections in z-axis obtained by 0.5 μm intervals confirm the localization of dystrophin signals, where 3-D stack of these eight sections is shown in (f). Scale bars: 50 μm (a–c); 20 μm (d–f)

WJ cells grown on plastic. Thus, hUCS-MSCs that were kept in culture for numerous doublings continued to express this myofibroblast marker. Given that the intrinsic properties of these cells are close to skeletal/smooth muscle cells, our rationale was to test whether myogenic induction via MyoD expression is an efficient way to transdifferentiate them into skeletal myocytes/myotubes, which would be proven by the expression of specific proteins. To date, no attempt has been reported for hUCS-MSCs addressing the muscle cell differentiation apart from the few studies

using 5-AZC, which principally acts as a transcriptional activator of silent genes by demethylating the genome that non-specifically forces the cell to differentiate. Therefore, cells driven by 5-AZC could apparently shift into the myogenic differentiation pathway, which is dominant over the other mesenchymal differentiation cascades [35]. However, at the same time, the non-specific action of 5AZC could drive cells into a series of deleterious events such as carcinogenesis [36]. On the contrary, the transient expression of the master transcriptional regulator of the myogenic cascade has been proven to be more effective, and thus was suggested as a more physiological mechanism of initiating the myogenic differentiation compared to 5AZC treatment [37]. Moreover, myogenic conversion with MyoD expression is considered safer based on the fact that it rapidly induces p21 and p27 [38], thus arresting the cell cycle progression, as we demonstrated with a marked downregulation of Ki67 expression in the course of differentiation. In vivo correction of genetic defects requires prompt fusion of the reprogrammed cells to the muscle syncytium. β-catenin signaling is crucial to muscle differentiation [39, 40] as well as an important component of the fusion machinery. The observation of rim-shaped homophilic interactions between fusing cells and concomitant induction

Fig. 4 Western blot analysis of β-catenin, M-cadherin, NCAM, and desmin during myogenic conversion. The molecular weights of the above observed protein bands are 92, 110, 120, 55 kDa respectively. Upon induction, β-catenin expression starts from the 12th hour and gradually increases, whereas M-cadherin and NCAM are firstly expressed on the 3rd day. The expression of desmin is clearly detectable following the 2nd day

Stem Cell Rev and Rep (2010) 6:512–522

519

Fig. 5 The mRNA levels of MyoD, β-catenin, desmin, MyHC, and Ki-67 are assessed by a reverse-transcriptase coupled by semiquantitative PCR during the first four-day period after myogenic reprogramming (n=3 for each group). The relative expression of the transcripts at various time points (days) is normalized to 0 h control values to represent a relative fold change of expression (y axis). The error bars represent the standard deviation of the biological replicates. PCR products of MB and muscle-specific isoform of CPK-M are documented on agarose gel electrophoresis up to 10 days (M: crude muscle mRNA)

of M-cadherin expression starting at day 3 as well as the NCAM upregulation following induction of differentiation are typical in terminally differentiating and fusing myoblasts [41, 42]. Myogenic regulatory factors, on the other hand, such as MyoD, are self-inducing; thus, once the expression is granted, they maintain their own expression for the maintenance of the muscle differentiation program [43]. The viral vector used in this study contained a CMV promoter, which exposes the cells to a heavy load of MyoD. While the hUCS-MSCs tolerated the vector without any prominent apoptosis [8], too much MyoD might finalize the myogenic differentiation rapidly without an adequate time frame for fusion [37]. We have observed up to 85% of transfection efficiency and in fact, we have recently found that compared to human bone marrowderived MSCs, the hUCS-MSCs express almost one log higher amount of coxsackie virus and adeno virus receptors (CARs), which are required for adenoviral vector gene transfers (unpublished data). Therefore, we suggest that hUCS-MSCs are favorable to bone marrow MSCs, which are almost devoid of CARs [44]. Nevertheless, from the therapeutical point of view, the use of adenovirus vectors is limited in view of the known immunogenicity issues in mammals. Alternatively, we also tested the efficacy of

liposomal gene transfer technique, and achieved a preliminary 65–70% transfection ratio up to 5 days of expression, which is readily enough for myogenic conversion. Native hUCS-MSCs exhibited numerous cell-to-cell contacts with surrounding xenotypic myoblasts. While showing successful heterokaryons with primary rat myoblasts, we could not demonstrate clear-cut multinucleated heterokaryonic myotubes in co-culture experiments with C2C12 cells. One possible reason for this is the overproliferation of C2C12 cells in the differentiation medium rendering the cells to become overconfluent for observation. Although controversial on fusion formation, we tested the effect of various concentrations of dbcAMP, which regulates the protein A kinase pathway [45]. However, upon varying doses of dbcAMP treatment, we did not note any human nuclei contribution to the C2C12 derived myotubes. The block over the spontaneous myogenic differentiation could easily be overcome upon reprogramming with MyoD expression as demonstrated by incorporation into the myotubes formed by primary rat myoblasts. The complete conversion of the hUCS-MSCs to the myogenic lineage is characterized by the observation of a sequence of muscle-specific structural proteins, enzymes and components of the contractile machinery. Among these,

520

Stem Cell Rev and Rep (2010) 6:512–522

Fig. 6 Non-transfected native hUCS-MSCs were incubated with mouse embryonic C2C12 myoblast cell line (green nuclei in a and c) and with rat myoblasts (d–f). Differentiating C2C12 cells occasionally exhibit cell-to-cell contacts with hUCS-MSCs, which were labeled with an antibody raised exclusively against nuclei of human cells (red signal in b and c). All cells were counterstained with fluorescein phalloidin for the visualization of F-actin microfilaments

(green intracytoplasmic filaments in a and c). Note the end-to-end and side-to-side contact sites (arrowheads in a). Costaining of a multinucleated myotube (GFP signal in d) directly indicates that a hUCSMSC labeled with anti-human nuclei (arrowhead in e) fuses with rat myoblasts as labeled with DAPI nuclear dye forming a heterokaryon (arrows in f). Scale bars: 50 μm

α2-laminin (merosin) is the specific extracellular matrix protein exclusively expressed and secreted by the differentiating myoblasts. Interestingly enough, we noted α2laminin as delineating the cytoplasmic borders of the fusing cells. Likewise, the expression of desmin, as well as the presence of the two structural components of the sarcomeric α-actinin and MyHC further indicated the successful myogenic conversion. The early differentiation marker NCAM and the above-discussed components of the fusion machinery—M-cadherin and β-catenin—indicated that those converting hUCS-MSCs effectively acquired the skeletal muscle phenotype. We were able to maintain the transfected cells up to 2 weeks in culture conditions, where the mRNA expression of MB and the muscle-specific

isoform of CPK were upregulated later in the course of differentiation. Finally, the sub-sarcolemmal localization of the dystrophin expression clearly demonstrated that our reprogramming strategy is capable of forming multinucleated muscle cells from hUCS-MSCs and can be used for the restoration of the dystrophin protein. Despite vigorous research over the last 10 years, DMD caused by the lack of the structural muscle protein dystrophin [46] is still an intractable disorder. There has been no successful clinical application of dystrophin gene replacement [47]. Currently, exon-skipping strategies to restore the reading frame are promising [48, 49] and clinical trials are ongoing with morpholino structures [50]. As the muscle is a syncytium of myocytes, long-term restoration of the genetic defect is

Stem Cell Rev and Rep (2010) 6:512–522

possible with a heterokaryon bringing in an intact genome. This heterokaryon concept also forms the basis of cell therapy approaches for other muscle dystrophies [51]. The search for an ideal transplantable nucleus donor is still ongoing and once proven to be effective; this approach would be applicable to all other inherited muscle diseases. To date, MSCs, bone marrow-derived progenitors and side population cells have been tested to achieve this goal. Myogenic induction of embryonic stem cell lines [52] and induced pluripotent stem (iPS) cells [53] are increasingly drawing attention for cell replacement therapy, but their liability for chromosomal anomalies and the need of lentiviral vectors, respectively, limit their therapeutical uses. While the mesoangioblasts were found to be the most promising cells capable of dystrophin restoration and incorporation into the muscle regeneration process [19], one important advantage of hUCS-MSCs over the above is the availability of a higher number of cells for transplantation without any need for in vitro expansion that has been proven to reduce the engraftment capacity in the case of myoblasts [54]. Secondly, hUCS-MSCs are devoid of MHC class II and low class I antigens, allowing them to be used for allogeneic or even xenotransplantations [55], which may not be possible for mesoangioblasts. Conclusively, our results indicate that hUCS-MSCs are exceptional candidates for gene delivery for muscle regenerative purposes. Acknowledgments The authors would like to thank Dr. Fadil Kara for generously providing human umbilical cords obtained from cesarean sections; to Genethon, Evry, France for kind supply of the adenoviral vectors. This work was partly supported by the Turkish Ministry of Industry (STZ-139-2007-2) to D. Balcı and A. Can and by the Scientific and Technological Research Council of Turkey (SBAG105 S364) to C. Kocaefe.

521

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

References 20. 1. Can, A., & Karahuseyinoglu, S. (2007). Concise review: human umbilical cord stroma with regard to the source of fetus-derived stem cells. Stem Cells, 25(11), 2886–2895. 2. Lu, L. L., Liu, Y. J., Yang, S. G., et al. (2006). Isolation and characterization of human umbilical cord mesenchymal stem cells with hematopoiesis-supportive function and other potentials. Haematologica, 91(8), 1017–1026. 3. Karahuseyinoglu, S., Cinar, O., Kilic, E., et al. (2007). Biology of stem cells in human umbilical cord stroma: in situ and in vitro surveys. Stem Cells, 25(2), 319–331. 4. Cho, P. S., Messina, D. J., Hirsh, E. L., et al. (2008). Immunogenicity of umbilical cord tissue derived cells. Blood, 111(1), 430–438. 5. Jomura, S., Uy, M., Mitchell, K., et al. (2007). Potential treatment of cerebral global ischemia with Oct-4+ umbilical cord matrix cells. Stem Cells, 25(1), 98–106. 6. Chao, K. C., Chao, K. F., Fu, Y. S., et al. (2008). Islet-like clusters derived from mesenchymal stem cells in Wharton’s Jelly of the

21.

22.

23.

24.

human umbilical cord for transplantation to control type 1 diabetes. PLoS ONE, 3(1), e1451. Liu, M., & Han, Z. C. (2008). Mesenchymal stem cells: biology and clinical potential in type 1 diabetes therapy. Journal of Cellular and Molecular Medicine, 12(4), 1155–1168. Rachakatla, R. S., Pyle, M. M., Ayuzawa, R., et al. (2008). Combination treatment of human umbilical cord matrix stem cellbased interferon-beta gene therapy and 5-fluorouracil significantly reduces growth of metastatic human breast cancer in SCID mouse lungs. Cancer Investigation, 26(7), 662–670. Ayuzawa, R., Doi, C., Rachakatla, R. S., et al. (2009). Naive human umbilical cord matrix derived stem cells significantly attenuate growth of human breast cancer cells in vitro and in vivo. Cancer Letters, 280(1), 31–37. Cao, F. J., & Feng, S. Q. (2009). Human umbilical cord mesenchymal stem cells and the treatment of spinal cord injury. Chinese Medical Journal (Engl), 122(2), 225–231. Liao, W., Zhong, J., Yu, J., et al. (2009). Therapeutic benefit of human umbilical cord derived mesenchymal stromal cells in intracerebral hemorrhage rat: implications of anti-inflammation and angiogenesis. Cellular Physiology and Biochemistry, 24(3–4), 307–316. Tsai, P. C., Fu, T. W., Chen, Y. M., et al. (2009). The therapeutic potential of human umbilical mesenchymal stem cells from Wharton’s jelly in the treatment of rat liver fibrosis. Liver Transplant, 15(5), 484–495. Blau, H. M., Chiu, C. P., & Webster, C. (1983). Cytoplasmic activation of human nuclear genes in stable heterocaryons. Cell, 32(4), 1171–1180. Terranova, R., Pereira, C. F., Du Roure, C., et al. (2006). Acquisition and extinction of gene expression programs are separable events in heterokaryon reprogramming. Journal of Cell Science, 119(Pt 10), 2065–2072. Partridge, T. A., Grounds, M., & Sloper, J. C. (1978). Evidence of fusion between host and donor myoblasts in skeletal muscle grafts. Nature, 273(5660), 306–308. Skuk, D., & Tremblay, J. P. (2003). Myoblast transplantation: the current status of a potential therapeutic tool for myopathies. Journal of Muscle Research and Cell Motility, 24(4–6), 285–300. Dreyfus, P. A., Chretien, F., Chazaud, B., et al. (2004). Adult bone marrow-derived stem cells in muscle connective tissue and satellite cell niches. The American Journal of Pathology, 164(3), 773–779. Dezawa, M., Ishikawa, H., Itokazu, Y., et al. (2005). Bone marrow stromal cells generate muscle cells and repair muscle degeneration. Science, 309(5732), 314–317. Sampaolesi, M., Blot, S., D’Antona, G., et al. (2006). Mesoangioblast stem cells ameliorate muscle function in dystrophic dogs. Nature, 444(7119), 574–579. Conconi, M. T., Burra, P., Di Liddo, R., et al. (2006). CD105(+) cells from Wharton’s jelly show in vitro and in vivo myogenic differentiative potential. International Journal of Molecular Medicine, 18(6), 1089–1096. Wang, H. S., Hung, S. C., Peng, S. T., et al. (2004). Mesenchymal stem cells in the Wharton’s jelly of the human umbilical cord. Stem Cells, 22(7), 1330–1337. Kadivar, M., Khatami, S., Mortazavi, Y., et al. (2006). In vitro cardiomyogenic potential of human umbilical vein-derived mesenchymal stem cells. Biochemical and Biophysical Research Communications, 340(2), 639–647. Pereira, W. C., Khushnooma, I., Madkaikar, M., et al. (2008). Reproducible methodology for the isolation of mesenchymal stem cells from human umbilical cord and its potential for cardiomyocyte generation. Journal of Tissue Engineering and Regenerative Medicine, 2(7), 394–399. Wu, K. H., Mo, X. M., Zhou, B., et al. (2009). Cardiac potential of stem cells from whole human umbilical cord tissue. Journal of Cellular Biochemistry, 107(5), 926–932.

522 25. Crouzet, J., Naudin, L., Orsini, C., et al. (1997). Recombinational construction in Escherichia coli of infectious adenoviral genomes. Proceedings of the National Academy of Sciences of the United States of America, 94(4), 1414–1419. 26. Yaffe, D., & Saxel, O. (1977). A myogenic cell line with altered serum requirements for differentiation. Differentiation, 7(3), 159–166. 27. Chomczynski, P., & Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry, 162(1), 156–159. 28. Vandesompele, J., De Preter, K., Pattyn, F., et al. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology, 3 (7), RESEARCH0034. 29. Takechi, K., Kuwabara, Y., & Mizuno, M. (1993). Ultrastructural and immunohistochemical studies of Wharton’s jelly umbilical cord cells. Placenta, 14(2), 235–245. 30. Eyden, B. P., Ponting, J., Davies, H., et al. (1994). Defining the myofibroblast: normal tissues, with special reference to the stromal cells of Wharton’s jelly in human umbilical cord. Journal of Submicroscopic Cytology and Pathology, 26(3), 347–355. 31. Chou, Y. H., Skalli, O., & Goldman, R. D. (1997). Intermediate filaments and cytoplasmic networking: new connections and more functions. Current Opinion in Cell Biology, 9(1), 49–53. 32. Nanaev, A. K., Kohnen, G., Milovanov, A. P., et al. (1997). Stromal differentiation and architecture of the human umbilical cord. Placenta, 18(1), 53–64. 33. Kobayashi, K., Kubota, T., & Aso, T. (1998). Study on myofibroblast differentiation in the stromal cells of Wharton’s jelly: expression and localization of alpha-smooth muscle actin. Early Human Development, 51(3), 223–233. 34. Mitchell, K. E., Weiss, M. L., Mitchell, B. M., et al. (2003). Matrix cells from Wharton’s jelly form neurons and glia. Stem Cells, 21(1), 50–60. 35. Konieczny, S. F., & Emerson, C. P., Jr. (1984). 5-Azacytidine induction of stable mesodermal stem cell lineages from 10T1/2 cells: evidence for regulatory genes controlling determination. Cell, 38(3), 791–800. 36. Carr, B. I., Rahbar, S., Asmeron, Y., et al. (1988). Carcinogenicity and haemoglobin synthesis induction by cytidine analogues. British Journal of Cancer, 57(4), 395–402. 37. Lattanzi, L., Salvatori, G., Coletta, M., et al. (1998). High efficiency myogenic conversion of human fibroblasts by adenoviral vector-mediated MyoD gene transfer. An alternative strategy for ex vivo gene therapy of primary myopathies. Journal of Clinical Investigation, 101(10), 2119–2128. 38. Halevy, O., Novitch, B. G., Spicer, D. B., et al. (1995). Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science, 267(5200), 1018–1021. 39. Borello, U., Coletta, M., Tajbakhsh, S., et al. (1999). Transplacental delivery of the Wnt antagonist Frzb1 inhibits development of caudal paraxial mesoderm and skeletal myogenesis in mouse embryos. Development, 126(19), 4247–4255. 40. Cossu, G., & Borello, U. (1999). Wnt signaling and the activation of myogenesis in mammals. The EMBO Journal, 18(24), 6867–6872.

Stem Cell Rev and Rep (2010) 6:512–522 41. Rose, O., Rohwedel, J., Reinhardt, S., et al. (1994). Expression of M-cadherin protein in myogenic cells during prenatal mouse development and differentiation of embryonic stem cells in culture. Developmental Dynamics, 201(3), 245–259. 42. Kramerova, I., Kudryashova, E., Wu, B., et al. (2006). Regulation of the M-cadherin-beta-catenin complex by calpain 3 during terminal stages of myogenic differentiation. Molecular and Cellular Biology, 26(22), 8437–8447. 43. Johnson, P., & Andrews, P. (2006). Cell fusion and the differentiation state. In R. Lanza (Ed.), Essentials of stem cell biology (pp. 81–85). Burlington: Elsevier Academic Press. 44. Kawabata, K., Sakurai, F., Koizumi, N., et al. (2006). Adenovirus vector-mediated gene transfer into stem cells. Molecular Pharmaceutics, 3(2), 95–103. 45. Shin, Y. J., Woo, J. H., Chung, C. H., et al. (2000). Retinoic acid and its geometrical isomers block both growth and fusion of L6 myoblasts by modulating the expression of protein kinase A. Molecules and Cells, 10(2), 162–168. 46. Koenig, M., Hoffman, E. P., Bertelson, C. J., et al. (1987). Complete cloning of the Duchenne muscular dystrophy (DMD) cDNA and preliminary genomic organization of the DMD gene in normal and affected individuals. Cell, 50(3), 509–517. 47. Romero, N. B., Braun, S., Benveniste, O., et al. (2004). Phase I study of dystrophin plasmid-based gene therapy in Duchenne/ Becker muscular dystrophy. Human Gene Therapy, 15(11), 1065– 1076. 48. Goyenvalle, A., Vulin, A., Fougerousse, F., et al. (2004). Rescue of dystrophic muscle through U7 snRNA-mediated exon skipping. Science, 306(5702), 1796–1799. 49. Goyenvalle, A., Babbs, A., van Ommen, G. J., et al. (2009). Enhanced exon-skipping induced by U7 snRNA carrying a splicing silencer sequence: promising tool for DMD therapy. Molecular Therapy, 17(7), 1234–1240. 50. Kinali, M., Arechavala-Gomeza, V., Feng, L., et al. (2009). Local restoration of dystrophin expression with the morpholino oligomer AVI-4658 in Duchenne muscular dystrophy: a single-blind, placebo-controlled, dose-escalation, proof-of-concept study. Lancet Neurology, 8(10), 918–928. 51. Mendell, J. R., Kissel, J. T., Amato, A. A., et al. (1995). Myoblast transfer in the treatment of Duchenne’s muscular dystrophy. The New England Journal of Medicine, 333(13), 832–838. 52. Darabi, R., Gehlbach, K., Bachoo, R. M., et al. (2008). Functional skeletal muscle regeneration from differentiating embryonic stem cells. Natural Medicines, 14(2), 134–143. 53. Takahashi, K., Tanabe, K., Ohnuki, M., et al. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131(5), 861–872. 54. Webster, C., & Blau, H. M. (1990). Accelerated age-related decline in replicative life-span of Duchenne muscular dystrophy myoblasts: implications for cell and gene therapy. Somatic Cell and Molecular Genetics, 16(6), 557–565. 55. Weiss, M. L., Anderson, C., Medicetty, S., et al. (2008). Immune properties of human umbilical cord Wharton’s jelly-derived cells. Stem Cells, 26(11), 2865–2874.