Yolk Sac Mesenchymal Progenitor Cells from New World Mice (Necromys lasiurus) with Multipotent Differential Potential

Fetal membranes are abundant, ethically acceptable and readily accessible sources of stem cells. In particular, the yolk sac is a source of cell lineages that do not express MHCs and are mainly free from immunological incompatibles when transferred to a recipient. Although data are available especially for hematopoietic stem cells in mice and human, whereas other cell types and species are dramatically underrepresented. Here we studied the nature and differentiation potential of yolk sac derived mesenchymal stem cells from a New World mouse, Necromys lasiurus. Explants from mid-gestation were cultured in DMEM-High glucose medium with 10% defined fetal bovine serum. The cells were characterized by standard methods including immunophenotyping by fluorescence and flow cytometry, growth and differentiation potential and tumorigenicity assays. The first adherent cells were observed after 7 days of cell culture and included small, elongated fibroblast-like cells (92.13%) and large, round epithelial-like cells with centrally located nuclei (6.5%). Only the fibroblast-like cells survived the first passages. They were positive to markers for mesenchymal stem cells (Stro-1, CD90, CD105, CD73) and pluripotency (Oct3/4, Nanog) as well as precursors of hematopoietic stem cells (CD117). In differentiation assays, they were classified as a multipotent lineage, because they differentiated into osteogenic, adipogenic, and chondrogenic lineages and, finally, they did not develop tumors. In conclusion, mesenchymal progenitor cells with multipotent differentiation potential and sufficient growth and proliferation abilities were able to be obtained from Necromys yolk sacs, therefore, we inferred that these cells may be promising for a wide range of applications in regenerative medicine.


Introduction
Fetal membranes are abundant, ethically acceptable and readily accessible sources of stem cells in contrast to the common albeit ethically questioned usage of embryonic stem cells [1][2][3]. Compared to stem cells derived from adult tissues, the fetal membrane cell lineages have greater differentiation potential and higher proliferation abilities. Also, they do not express Major Histocompatibility Complex (MHC) surface markers and thus do not cause immunological incompatibilities when transferred to a recipient [4,5]. Special attention has been drawn to the yolk sac, because this membrane maintains essential functions during early gestation including the presence of pluripotente primordial germ cells that subsequently migrate to the primitive gonads, it contribute to form the midgut as well as the first hematopoietic stem cells that contribute to the development of the vascular system [6][7][8][9]. In addition, the yolk sac endoderm promotes maternal-fetal exchange [10]. So far, much data on cell differentiation from yolk sac tissues are available for hematopoietic stem cells from mice and humans [6,8,[11][12][13][14][15], whereas other cell types and species are dramatically underrepresented. This is especially the case for mesenchymal stem cells that are regarded multipotent at least. Mesenchymal stem cells from embryos and adults have been recognized for their ability to differentiate in vitro and in vivo into osteogenic, adipogenic and chondrogenic cell lineages [16,17]. In cell culture, mesenchymal cells have a fibroblast-like phenotype [18]. According to The International Society for cellular therapy position statement, human mesenchymal stem cells can be clearly identified by the characteristic responses to a set of cell type, differentiation and potency markers, i.e. positive response to CD105, CD73 and CD90 and negative response to CD45, CD34, CD14 or CD11b, CD79a, CD19 and HLA [19]. However, the limited proliferation and differentiation potential of adult mesenchymal stem cells [20] and the risk of tumors and ethical issues when using embryonic tissues were the impetus to continue research using fetal membranes. Indeed, the yolk sac was recently reported as a source for mesenchymal stem cells. Such cells from the murine yolk sac differentiate into osteoblasts, chondrocytes and adipocytes [21,22]. For the human, yolk sac mesenchymal stem cells differentiated into osteogenic, adipogenic and neurogenic lineages [23][24][25], but not into chondrocytes [25]. In addition to the above characteristics, both rodent and human yolk sac derived cells are also immunopositive for further antigens as well as markers for proliferation and pluripotency [22][23][24][25]. This pattern, and differentiation into mesoderm-derived lineages only are regarded as indications of multipotency. Finally, although there were promising initial results on other species such as dog [26] or sheep [27], a broader sample of species is desired. In particular, it regarding development of animal models to determine whether the mouse reflects a more specialized or more general pattern of rodents. Thus, here we studied the nature and differentiation potential of yolk sac-derived mesenchymal progenitor cells from a New World mice species native to South America, Necromys lasiurus (Rodentia, Cricetidae, Sigmodontinae). Explants from mid-gestation yolk sacs were cultured and thereafter cells were characterized by standard methods including immunophenotyping by fluorescence and flow cytometry to identify surface antigen expression, growing and differentiation performance and tumorigenicity assays.

Collection, Cell Culture and Cell Morphology
Yolk sacs of Necromys lasiurus from 10 individuals in midgestation (15-16 days) were obtained from a breeding group at the University of Mossoró and cultured at the University of Sao Paulo. The samples were plated in 35 mm Petri dishes (Corning, NY, USA) with four media ( Table 1). The Petri dishes were incubated at 37uC in a humidified atmosphere of 5% CO 2 . After 24 and 48 hours, no-adherent cells were removed and the medium was replaced. Every 3 days, 70% of the medium was replaced and at 80% of confluence, the cells were harvested with 0.25% trypsin solution (Invitrogen, Carlsbad, CA, USA) and replaced in 25 cm 2 and 75 cm 2 flasks (Corning, NY, USA). Progenitor cells from the yolk sac were analyzed morphologically every 3 days, using an inverted microscope (NIKON ECLIPSE TS-100) and the size and format of the cell populations were also evaluated by flow cytometry (FACSCalibur, BD, San Jose, California, USA). The immunophenotype characterization of cells was done on passage four.

Colorimetric Assay (MTT)
The colorimetric [3-(4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT)] assay was performed according to the protocol previously established [28]. In order to verify cellular proliferation, the analysis was performed every 48 hours over a 12day. For this, 1610 5 cells were plated on 96 wells plates. Basal culture medium was replaced concurrent with analysis. The cells were washed with 100 mL PBS/well; then, 100 ml of MTT solution [1 mL of MTT (5 mg/mL) in 10 mL of culture medium, and 1% of fetal bovine serum] was added and maintained at 37uC for 3 hours. Thereafter, the formazan was precipited and

Immunophenotyping by Flow Cytometry
Yolk sac cells on passage 4 were re-suspended in FACS (fluorescence-activated cell sorting) buffer, and the concentration adjusted to 10 5 cells/mL. For intracytoplasmic and nuclear markers, cells were permeabilized with 5 ml 0.1% Triton X-100 for 30 minutes prior to incubation with specific primary antibodies

Adipogenic, Osteogenic and Chondrogenic Differentiation Assays
All differentiation assays were performed in triplicate using yolk sac cells on passage 6. For osteogenic and chondrogenic differentiation, 1610 4 yolk sac cells were plated in six-well plates with a basal culture medium. After 24 hours, osteogenic differentiation was induced with a medium composed of DMEN-High, 7.5% of fetal bovine serum, 100 mM ascorbate The medium was changed every 3 days. At day 21, the cells were washed twice in PBS and fixed for 24 hours in 4% paraformaldehyde. To detect intracellular lipid accumulations, cells were stained using Sudan II (SC-215923, Santa Cruz, CA, USA) for 5 minutes, washed in 70% alcohol, and in Harris haematoxilin for 2 minutes. The chondrogenic differentiation was done according to protocol established by Mambelli et al. [29]. To prepare a pellet culture, 2610 6 cells were re-suspended using a commercial medium (STEMPRO Chondrogenesis Differentiation Kit, Invitrogen, Carlsbad, CA, USA). Half of the medium was changed twice per week over 21 days. During the culture interval, cells were maintained at 37uC in a humidified atmosphere with 5% CO 2. After 21 days, cell aggregates were embedded in paraffin and sectioned at 3-4 mm in an automatic microtome (LEICA, RM2165). Then, slides were stained using Picrosirius and Masson's trichrome.

Tumorigenicity Assay in Nude Mice
Progenitor yolk sac cells (n = 1610 6 cells) on passage 4 were resuspended in DPBS solution (Invitrogen) and intramuscular injected into the left limbs of 2 nude mice. The animals were maintained under specific pathogen-free conditions using an Isorack, and fed on sterile food and water ad libidum at the IPEN (Nuclear and Energy Research Institute, University of Sao Paulo, Brazil) for 8 weeks. Every week, the animals were clinically examined to identify possible tumor formation. Then, the animals were euthanazied following the principles of the Ethical Committee of the School of Veterinary Medicine and Animal Science and samples from the biceps, liver, lung, kidney, and abdominal adipose tissue were collected and fixed in 4% paraformaldehyde. Tissues for histopathology were embedded in paraffin and sectioned at 5 mm, stained with haematoxylin and eosin (HE) and investigated with an Olympus microscope (CX 31 RBSFA).

Isolation and Growth Characteristics of the Yolk Sac Cells
After tested four culture media (Table 1), the best results in regard to cell growth and proliferation were obtained using a combination of DMEM-High glucose medium supplemented with 10% fetal bovine serum defined. Thus, this medium was chosen to perform the experiments. In the primary cell culture the first adherent cells were observed 7 days after the explants were plated. They included two types of cells: fibroblast-like cells were small and elongated and had reduced cytoplasm, whereas epithelial-like cells were large and rounded with sparse cytoplasm ( Figure 1A). Both types had centrally located nuclei. Only the fibroblast-like cells survived continued cell passages and application of freezing assays; still showing satisfactory growth potential and uniformity ( Figure 1B). In addition, fibroblasts were much more abundant in the cultures than epithelial-like cells (92.1% versus 6.5%; Figure 1C). They were confluent in the plates and maintained the above-described morphology. In order to understand the cellular proliferation by the colorimetric assay, the 1610 5 fibroblast-like cells had progressive growth until day 5. After this, was observed a decrease of 22% in cell viability ( Figure 1D).
Usually, the yolk sac cells especially the fibroblast-like cells remained growing satisfactorily until passage 4 where the culture was more homogeneous than the previous passages. For this reason, the experiments were performed on passages 4 and 6.

Immunophenotype Cell Characterization
The expression of specific cell types, proliferation and differentiation markers were analyzed by a combination of immunofluorescence and flow cytometry using cells on passage four ( Figures 1E-H, 2,3). In both methods, fibroblast-like cells were immunopositive for vimentin ( Figure 1E

Differentiation Ability
On passage 6, the yolk sac derived cells were induced to adipogenic, osteogenic, and chondrogenic differentiation by applying specific media routinely used for mesenchymal stem cell differentiation. After 21 days, cells differentiated into the three lineages. Cells with adipocyte morphology were organized in small groups. Their nuclei were displaced and closely attached to the cell membranes. Intracellular lipid accumulations were observed in the cytoplasm after Sudan II staining ( Figure 4A). Osteogenic differentiation was confirmed with Von-Kossa staining and by detection of alkaline phosphatase activity. Then the cells were polygonal and had a mineralized extracellular matrix ( Figure 4B), and they were able to produce alkaline phosphatase enzyme ( Figure 4C). Under chondrogenic differentiation, a suspended pellet formation occurred. Based on Masson trichrome staining, the cells with condrocyte lacunae were frequent ( Figure 4D). The matrix was composed of extended collagen areas that were rich in collage fibers ( Figure 4E). Using Picrosirius staining under polarized light, collagen areas were quantified (Image J Program) to occupy 42.3% of the total sample ( Figure 4F), whereas the control cells without treatment remained their undifferentiated morphology.

Tumorigenicity of the Yolk Sac Cells
Immunosuppressed or nude mice did not have any indication of tumor formation within 8 weeks after injection of the yolk sacderived mesenchymal progenitor cells, neither macroscopically ( Figure 5A) nor histopathologically in samples of lung ( Figure 5B), skeletal musculature ( Figure 5C), liver ( Figure 5D), kidney ( Figure 5E), and adipose tissue of the abdominal region ( Figure 5F).

Discussion
Although only a few taxa have been investigated in that regard, former studies suggested that the yolk sac is a promising source for mesenchymal progenitor cells [25,26,30]. As typical for cells derived from fetal membranes, they do not express MHC on the surface and do not cause immunological incompatibilities usually associated with adult stem cells. In addition, extraembryonic tissues are ethically less problematic to use than parts of the embryo proper. The current results in a species of New World mice (Necromys lasiurus) supported these views. We were able to isolate, culture and characterize mesenchymal progenitor cells with defining reactions to immunological surface markers for cell type, differentiation status and proliferation. In addition, the cells had sufficient growth parameters and differentiated into osteogenic, adipogenic and chondrogenic lineages without developing tumors. Thus, the typical characteristics of mesenchymal stems cells from the mouse yolk sac seemed to be widespread among rodents or even larger clades of placental mammals including the human.
From the four tested culture media, the DMEM-High Glucose medium supplemented with defined fetal bovine serum had the best results regarding to cell adhesion, growth and maintenance of the undifferentiated morphology of the fibroblast-like cells during the passages. Interestingly, DMEM-High Glucose medium with characterized fetal bovine serum caused death of the cells. Thus, the type of fetal bovine serum seems to play an important role in the survival and growth of the yolk sac cells and may be it is related to the filtering process of these serums. In summary, the HyClone defined and characterized fetal bovine serums are filtered using 40 nm (0.04 mm) and 100 nm (0.1mm) pore-size filters, respectively. For mesenchymal stem cells from either human [25] or dog yolk sac [26] the authors demonstrated that ALPHA-MEM was the best medium to culture these cells, but they did not specify in detail regarding fetal bovine serum that was used in the culture. The culture of yolk sac explants resulted in a heterogeneous population that contained mainly fibroblast-like cells, but also a few epithelial-like cells. The latter must have been derived from the endothelial and mesothelium layers that characterized the inverted yolk sac of rodents including Necromys [31]; however, these cells died during the first cell passages. The fibroblast-like habits of the surviving cells [18] as well as the characteristic positive and negative responses to a set of surface markers clearly identified them as mesenchymal progenitor cells. This was apparently the first application of immunophenotyping markers in both flow cytometry and fluorescence, with consistent results. The cells were positive to mesenchymal stem cell markers (CD 73, 90, 105) and immunonegative to CD 34 and 45 as markers for hematopoetic stem cells. Thus, our cells had the characteristic pattern used to define mesenchymal stem cells in general [19]. They were also positive for Stro-1, a newer and probably better marker for such stem cells [32]. Interestingly, our fibroblast-like cells were immunopositive to CD 117 as a marker for haematopoetic cell precursors, suggesting that the cells still had the potential to give rise to hematopoietic lineages or vasculature components when using an induction medium. This was supported by the positive response to VEGF usually associated with development of the vascular system [33,34]. Moreover, it is important to consider the importance of mesenchymal cells for hematopoiesis [35]. In addition, the positive response to PCNA fitted well into the observed growth and proliferation pattern. A similar pattern was reported for human mesenchymal stem cells [25]. Also, the cells were immunopositive to vimentin, cytokeratin and b-tubulin as markers for the cytoskeleton [36] and a high differentiation potential was indicated by the expression of the pluripotency markers Oct-3/4 and Nanog typical for progenitor cells [9,37]. This pattern was consistent with previous results on yolk sac derived mesenchymal progenitor cells from the mouse, human and dog [22][23][24][25][26]. However, recent results from the human indicate that the exact effect of Oct4 is not fully understood [38], although it is regarded as a most critical regulator of pluripotency that regulates Nanog [39]. In addition, perhaps the differential potential of stem cells was due to a statistical property of cell populations instead of being exactly defined at the single-cell level [40].
The progenitor stem cells from the Necromys yolk sac were able to differentiated into adipogenic, osteogenic and chondrogenic lineages when using culture media with inducing factors [41]. A similar far-reaching differentiation potential was reported from mesenchymal progenitor cells of the yolk sac of the mouse [21,22]. However, mesenchymal progenitor cells from other species had more limited potency. Adipogenic and osteogenic lineages were reported for the human, although the cells failed to differentiate into the chondrogenic lineage [24,25]. Progenitor cells from the dog yolk sac differentiated into osteogenic and chondrogenic lineages only [26]. In contrast, human mesenchymal stem cells from adult tissues also had far-reaching differentiation potential [42]. Studies cited above indicate that the differentiation potential is not pluripotent and do not lead to cell lineages characteristic for all three germ layers, but to mesoderm-derived ones only, which is regarded as an indication for multipotency. Finally, based on the present and previous data [26] differentiation of the cell lineages from fetal membrane tissues such as the yolk sac required approximately 3 weeks, which was longer than differentiation assays derived from adult tissues [43]. Last but not least, our data confirmed that the cultured yolk sac derived progenitor cells did   not cause development of tumors when transferred to a recipient, similar to other species [21,22,[24][25][26].

Conclusion
Mesenchymal stem cells with multipotent differentiation potential and sufficient expansion abilities were obtained from Necromys yolk sacs. The typical characteristics of mesenchymal progenitor cells from the mouse yolk sac seemed to be widespread among rodents; therefore, we inferred that such cells have a wide range of applications in regenerative medicine.