Embryonic fibroblasts represent a connecting link between mesenchymal and embryonic stem cells


  • Conflict of interest: None declared.

Author to whom all correspondence should be addressed.

Email: rajarshi.pal@manipal.edu


It is well established that fibroblasts and mesenchymal stem cells (MSC) share several characteristics with subtle differences. However, no study highlighting the versatility of fibroblasts beyond their multipotentiality has been reported so far. Mouse embryonic fibroblasts (MEFs) are widely used as feeder layers to support the growth of embryonic stem cells (ESC). We hypothesized that MEF may retain ES-like features in concurrence to their developmental hierarchy in addition to their multipotent nature. Hence, we performed a comparative assessment of MEF and ESC to determine their ability to differentiate into cell types other than mesoderm as well as capacity to form teratoma using routine in vitro and in vivo techniques. MEF were derived by trypsin/ EDTA (ethylenediaminetetraacetic acid) digestion from E13.5 embryos after removing heads and viscera following plastic adherence. MEFs robustly proliferated in culture until passage 15 and formed aggregates by hanging drop method. Flow cytometry, reverse transcription–polymerase chain reaction (RT–PCR) and immunocytochemistry revealed the presence of key MSC markers such as CD90, CD73, Sca-1, CD44, CD29, Vimentin and absence of CD45. Additionally, they expressed SSEA-1, Oct-4, Nanog, Sox-2 and ABCG2 as pluripotency markers; Nestin, β-III tubulin, Otx-2 (ectoderm); MEF-2, Mesp2, GATA-2 (mesoderm) and GATA-4, α-amylase, PDX-1 (endoderm) as tri-lineage markers. Furthermore, MEFs formed representative tissues from all three germ layers upon transplantation into Balb/c mice. These unique abilities of MEF to exhibit pluripotency, in addition to fibroblast characteristics and their ready availability with less ethical concerns and low maintenance requirements make them an attractive model for further exploration as a possible tool for regenerative medicine.


Fibroblasts are found to exist in every organ of animals contributing to their self-renewal, tissue remodeling and repair, facilitating the functional maintenance of the organs. Generally they are plastic adherent, multipotent, non-endothelium, epithelium or hematopoietic in origin and have the capacity to synthesize and remodel the extracellular matrix. In addition to their supportive function, they are also involved in regulating self-tolerance, organ development, wound healing, inflammation and fibrosis, which make them an attractive in vitro model (Haniffa et al. 2009; Kuroda et al. 2010) for different kinds of studies both basic and translational. These fibroblasts have been widely used as feeder layers to support the growth and expansion of embryonic stem cells (ESCs); most commonly the mouse embryonic fibroblasts (MEFs) (Amit et al. 2004). The MEFs are derived from 12.5 to 13.5 days post coitum (dpc) after removing the head and viscera region, while the trunk and the mesodermal tissues are the major contributors of MEF. They are also used to study a number of biological properties such as cell cycle regulation, immortalization, transformation, senescence, apoptosis and differentiation (Harris et al. 1994; Lowe et al. 1994; Kamijo et al. 1997; Steinman et al. 2004; Horwitz et al. 2005; Dominici et al. 2006; Haniffa et al. 2007, 2009; Lysy et al. 2007; Sun et al. 2007).

As opposed to MEFs, which are poorly investigated; recently mesenchymal stem cells (MSCs) have been studied extensively and have emerged as a promising candidate for a number of applications in cell based therapies. These MSCs are multipotent, clonogenic, plastic adherent, non-tumorigenic and are easily available in different prenatal and postnatal tissues (Alt et al. 2011; Huang et al. 2012; Saeed et al. 2012) albeit in small numbers. In fact MSCs can be isolated from bone marrow (BM), skeletal muscle, pancreas, vessels, dental pulp and adipose tissue (Zuk et al. 2001; Jiang et al. 2002; Hu et al. 2003 ; Yoshimura et al. 2007; Crisan et al. 2008; Nakatsuka et al. 2010). They display a number of immunomodulatory and immunosuppressive properties that play a major role in wound healing and repair (Haniffa et al. 2009). However, MSCs have certain limitations for clinical applications as they are adult tissue-specific stem cells; show restricted proliferation and differentiation capacity. To overcome these shortcomings, pluripotent embryonic stem (ES), embryonic germ (EG) and induced pluripotent stem (iPS) cell lines were developed. The ESCs are isolated from the inner cell mass (ICM) of the embryo and have the unique capacity of unlimited self-renewal and plasticity, which means that they can give rise to any cell type of the body (Martin 1981). There are several features that define ESCs that include their pluripotent origin, ability to undergo infinite rapid expansion in their primitive immortal state with stable karyotype and to form teratomas under in vivo conditions. Although, they possess many advantages, some serious drawbacks of using ESCs in regenerative medicine are its immunogenicity, teratoma formation and ethical issues (Kuroda et al. 2010).

Under such circumstances, it is remarkable to note that in the recent years there has been tremendous interest among the scientists to hunt for a suitable cell type that can exhibit characteristics of both MSCs as well as ESCs. In this direction, several reports have conclusively shown that MEFs are similar to MSCs phenotypically, genotypically and functionally. Both MEFs and MSCs are defined as plastic adherent, colongenic, multipotent fibroblast like cells that have similar morphological appearance, identical cell surface markers: CD90, CD73, CD105, Vimentin and negative for hematopoietic markers CD14, CD34 and CD45 and the ability to undergo mesoderm-type cell differentiation into osteocytes, adipocytes and chondrocytes (Harris et al. 1994; Horwitz et al. 2005; Dominici et al. 2006; Haniffa et al. 2007, 2009; Lysy et al. 2007). However, to date no reports are available wherein the fibroblasts have been studied for pluripotent behavior beyond their established notion of being multipotent cells.

Being optimistic, we hypothesized that stem-like cells exist among the embryonic fibroblasts, which possess the ability to differentiate into multilineage cell types. Based on their developmental hierarchy, these cells might have inherent characteristics somewhat similar to pluripotent stem cells. Thus, in this study we investigated whether stem-like cells exist in MEF and their identity was confirmed by their gene and protein expression. The in vitro multi-lineage differentiation potential of MEFs was assessed by the formation of aggregates. The MEF aggregates were tested for expression of markers representing the three germ layer makers like ectoderm, mesoderm and endoderm. Further, in vivo differentiation capacity of MEFs was evaluated by teratoma formation assay in Balb/c mice.

Materials and methods

Cell culture of primary mouse embryonic fibroblasts

The required approval for isolation and culture of primary mouse embryonic fibroblasts (MEFs) was obtained from the Institutional Animal Ethics Committee (IAEC), Manipal University. The mice were induced for pregnancy at the Manipal University Animal House. The primary MEFs were isolated using the standard protocol (Hogan et al. 1994; Zhang et al. 2003). The E12.5–13.5 pregnant mice were killed and the uterine horns were removed. The embryos were removed and washed with phosphate-buffered saline (PBS) containing antibiotics. The embryo was further dissected by removing their placenta, heads, limbs and gonads, tail and other visceral mass. The cells were isolated both by mechanical activity (chopping the tissue into fine pieces) and enzymatic treatment (digestion of tissue with 0.25% Trypsin/ethylenediaminetetraacetic acid [EDTA]). The enzymatic activity was neutralized by adding MEF media containing Dulbecco's modified eagle medium (DMEM)-HG, 10% fetal bovine serum (FBS), 1% Penicillin-Streptomycin and 1% glutamine and the tissue was pipetted up and down to get a single cell suspension. The cells were cultured in T-75 flasks until 75–80% confluent and subcultured at a ratio of 1:3. In this study, MEFs were used at different passages (P1–10).

Mouse ES cell culture

R1 mouse ES cell line from P33 onwards was cultured following the standard protocol. Briefly, mES colonies were grown on a mitotically-C (Sigma) inactivated (10 μg/mL for 2.5 h) MEF in 35 or 60 mm tissue culture plates (Falcon) pre-coated with 0.1% gelatin. The medium consisted of DMEM-high glucose supplemented with 15% FBS (Hyclone), 1× non-essential amino acids, 2.0 mmol/L glutamine, 1000 U/mL mouse recombinant LIF, 100 μmol/L 2-mercaptoethanol, 100 U penicillin, and 100 μg/mL streptomycin (all from Life Technologies unless mentioned). The cells were passaged in 2–3 days depending on the size and morphology of the ES colonies using 0.05% Trypsin/EDTA. Cultures were fed everyday owing to its very high growth rate.

Cell proliferation study by population doubling

Proliferation of MEFs was assessed by population doubling. The MEFs at P6 were plated at a density of 10 000 cells/cm2 in a 24-well plate. These cells were cultured at 37°C, 5% CO2 in MEF media. After every 24 h, the cells were trypsinized by incubation with 0.25% trypsin/EDTA for 2 min at 37°C. The trypsin activity was neutralized with MEF media. The cell count after every 24 h was determined by trypan blue exclusion method and recorded. The population doubling (PD) time was extrapolated from the graph.

Cell cycle analysis by propidium iodide staining

Approximately 5 × 105 cells at P6 were trypsinized, collected and counted as mentioned earlier. These cells were fixed with ice cold ethanol (70%) and centrifuged at 10 000 g for 5 min. These cells were then washed once with ice cold PBS. The pellet was resuspended in propidium iodide (PI) (1 mg/mL) containing RNAase-A (20 mg/mL) and was incubated at room temperature. The distribution pattern of cells in G1, S, and G2 phases was determined using Cell Quest software.

β-Galactosidase staining senescence assay

The β-galactosidase stain (Cell Signaling Technology, USA) was used to assess the cellular senescence. The cells were plated at different passages (P3 and P9) in a 6-well plate and were cultured until it reached 70–80% confluency. The cells were washed twice with PBS and then fixed with a fixative solution (20% formaldehyde, 2% glutaraldehyde in 10× PBS) for 10–15 min at room temperature. The cells were again washed and then stained with β-galactosidase stain at 37°C, overnight in a dry incubator. The senescent cells appeared bluish green under the light microscope.

Cell surface marker analysis by flow cytometry

The cells from P3 to P4 were harvested using 0.25% Trypsin/EDTA and then fixed in 1% sodium azide and 2% paraformaldehyde in PBS. The phenotype analysis was performed using fluorescein isothiocyanate (FITC) and phycoerthrin (PE) conjugated antibodies against CD44, CD45, CD73, CD90 and Sca 1 as well as isotype matched controls. Flow cytometry was performed on live cells for data acquisition using a FACS Caliber (Becton Dickinson, USA) and analysis was done using CellQuestPro software.


Mouse embryonic fibroblasts at P8 were grown in monolayers in 4-well chamber slides and fixed with 4% paraformaldehyde. The cells were washed with 1 × PBS, permeabilized with 0.1% Triton X-100 (for nuclear proteins) and blocked using 0.1% bovine serum albumin (BSA) in PBS for 30 min. These cells were then incubated with primary antibodies (anti mouse/rabbit) Vimentin, SSEA-1, Ki-67, Nestin, α-smooth muscle actin (SMA), PDX-1 and C-peptide overnight at 4°C followed by their respective secondary antibodies for 1.5 h in the dark. The cells were then counterstained with DAPI (4, 6-diamidino-2-phenylindole) (1:1000 dilution) and mounted using antifade mounting medium containing (DABCO [1, 4-diazabicyclo 2.2.2] octane). Finally slides were observed using fluorescence microscope (Olympus) and images were captured from different fields.

Adipocyte differentiation

The MEFs were plated at 8 × 103 cells/cm2 in 24-well plates in KO-DMEM media containing 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin. When cells reached 95% confluence, media was changed to adipocyte induction medium (AIM) containing: 5 μg/mL Insulin, 1 × 10−6 mol/L dexamethasone, and 0.5 μmol/L 3-isobutyl-1-methylxanthine (IBMX). The media was changed every 2nd day for 15 days. For characterization, cells were fixed with 10% formalin for 20 min at room temperature, rinsed with 60% isopropanol solution, and stained with Oil Red-O (Sigma) solution for one hour at room temperature and then the stain was removed by rinsing with distilled water. The cultures were observed under phase contrast microscope and images were captured.

Osteoblast differentiation

The MEFs were plated at a density of 8 × 103 cells/cm2 in the presence of KO-DMEM, 10% FBS and 1% P/S. At 70% confluency, cells were induced with 100 μg/mL Vitamin C and 10 mmol/L β-glycerol-phosphate. The media was changed every 2nd day for 15 days. For characterization, cells were fixed with 10% formalin for 20 min at room temperature and rinsed with distilled water; 5% AgNO3 was added and exposed to UV light for 30 min. Then 2.5% sodium thiosulphate was added for 5 min at room temperature; finally the stain was removed by rinsing with distilled water. The cultures were observed under phase contrast microscope and images were captured.

Formation of aggregates/spheroids

The MEFs were induced to form aggregates by hanging drop method followed by plating the small clusters in a non-adherent glass dish containing serum free media. For hanging drop method, about 20 μL of cell suspension was seeded on the lid of the dish and the base of the dish was plated with media, so as to maintain the humidity. The cell aggregates or spheroids were cultured in this media for 4 days. On the 4th day the media was removed and the cells were harvested for RNA isolation and immunocytochemistry.

Gene expression analysis by RT–PCR

Cells were lysed with Trizol Reagent (Life Technologies, USA); RNA was extracted according to the manufacturer's protocol and quantified using a Nanodrop spectrophotometer. RT was performed in a final volume of 20 μL with 1 μg total RNA containing 200 U of Superscript III Reverse Transcriptase. PCR primers specific for CD29, CD44, SCA-1, α-SMA, Vimentin, Oct-4, Nanog, Sox-2, ABCG2, Nestin, β-III tubulin, Otx-2, MEF-2, Mesp2, GATA-2, GATA-4, α-amylase and PDX1 were used in 10 mmol/L concentration. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene control. The PCR was run on a gradient thermal cycler (Eppendorf) for 35 cycles consisting of 97°C (5 min), 96°C (30 s), respective annealing temperature (45 s), 72°C (30 s); followed by final extension for 10 min at 72°C. The PCR products were then analyzed by electrophoresis in 2% agarose gels and images were clicked in a gel documentation system (BioRad, USA).

In vivo studies by teratoma formation assay

Male Balb/c mice of 3–4 weeks old weighing 25–30 g were used for in vivo studies. All animal experiments were performed in accordance with Institutional Ethical Committee for Animal Experiments and Committee for the Purpose of Control and Supervision on Experiments on Animals (CPCSEA) guidelines and regulations. The mice were immunosuppressed by cyclosporine injection (1 mg/kg body weight) and then MEF transplantation was carried out. One and half a million MEF were washed, resuspended in DMEM complete medium and then mixed with prewarmed 1.5% agarose to entrap cells in molten agarose. The cells were then injected subcutaneously into Balb/c mice (maintained in in-house animal house facility) in 1.5% melted agarose. The left side agarose plug was without cells and right side agarose plug was with MEF cells. Visible tumors appeared after 2 weeks of transplantation. These tumors were dissected out and fixed overnight with 4% PFA solution. The tissues were then paraffin embedded, sectioned, stained with H and E, and examined for the presence of tissue representatives of all three germ layers produced by MEF. The same number of undifferentiated mouse ES cells at P37 was used injected as a positive control for teratoma formation assay.

Statistical analysis

Statistical analysis was performed using a two-tailed unpaired Student's t-test. Differences are considered statically significant when < 0.05.


Isolation, growth kinetics and surface marker expression of embryonic fibroblasts

The MEFs isolated from E12.5 to 13.5 days of pregnancy could be maintained in culture consistently for more than 10 passages. These MEFs exhibited higher proliferation rates and shorter population doubling time of about 24 h (Fig. 1A). However they tended to become more flattened in morphology towards the later passages such as P9–P11 with increase in doubling times (data not shown). The MEFs at P11 showed senescence as indicated by bluish green cells in β-galactosidase assay when compared to unstained MEF at P5 (Fig. 1B–D). Our results are in similar line with a recent study by Kassem and coworkers in 2012, where MEFs exhibited a higher cell proliferation rate in short-term cultures, and a shorter population doubling time (PDT) of 3.43 days. We reason that the difference in PDT could be due to the difference in culture conditions. The growth kinetics of MEFs was determined by staining approximately 1.34 million cells with PI. The histogram showed the highest quiescent G0/G1 population (66%) compared to mitotic G2/M phase population (28%) and synthesis S phase population (5%), suggesting that these cells have a propensity to either maintain the quiescent stage or undergo cell division, promoting its multipotent characteristics (Fig. 1E,F). Taken together, these results demonstrate that MEFs show a higher proliferative potential in vitro.

Figure 1.

Isolation and growth kinetics of mouse embryonic fibroblasts (MEFs) (A) The population doubling time (PDT): 24 h and the fibroblastic morphology of MEFs at P1 day 2 (×4 magnification). (B–D) β-Galactosidase staining for MEFs at P5 and P11 (×4 and ×20 magnification); enlarged view of β-gal+ cells in MEF (P11) cultures. (E–F) Dot plot and histogram of cell cycle analysis by PI staining in MEFs at P5. The histogram represents the G0/G1 phase (66%), G2/M phase (28%) and S phase (5%), respectively.

The immunophenotyping studies confirmed that the MEFs were positive for MSC markers such as CD90 (92%), Sca-1 (94%) and CD44 (60%) and negative for hematopoietic marker CD45 (1.7%) (Fig. 2A–F). The surface antigen expression profile conformed to the criteria generally defined for mouse multipotent MSCs (Dominici et al. 2006). These results demonstrate that the MEFs express most if not all of the conventional MSC markers, suggesting that these cells may possess MSC-like phenotypes.

Figure 2.

Flow cytometric analysis of cell surface marker expression in mouse embryonic fibroblasts (MEFs) at P5. The MEFs are largely positive for MSC markers: CD90 (92%), Sca-1 (94%), CD44 (66%) and negative for CD45 (1.7%) in comparison with appropriate isotype controls. Data are representative of three independent experiments.

Embryonic fibroblasts express mesenchymal markers and differentiate into adipocytes and osteoblasts

Mouse embryonic fibroblasts clearly showed a higher expression of MSC markers such as CD29, CD44, Sca-1, α-SMA and Vimentin as compared to MEF aggregates (Fig. 3A). Mouse ESC was taken as a control where the mesenchymal markers were either absent or downregulated significantly (Fig. 3A). The immunocytochemisty data of MEFs also confirmed the expression of Vimentin, an MSC cytoskeletal marker (Fig. 3B). Further, the MEFs upon induction readily differentiated into large number of adipocytes and to a lesser extent to osteocytes as evidenced by positive staining with Oil Red-O and Von Kossa, respectively (Fig. 3C,D). We assume that the lack of BMP-2 in our induction media could account for the lower osteogenic differentiation as BMP-2 has been reported to be essential to induce osteogenesis in fibroblasts (Rutherford et al. 2002). The MEFs also showed spontaneous differentiation into neural progenitor-like cells (Fig. 3E). The phase contrast microscope pictures showed that the MEFs exhibited a triangular morphology with several processes or projections similar to the neural phenotype (Fig. 3E inset). As the MEFs were isolated from the trunk region of the embryo, it is not surprising that these cells may share similar properties to neural cells, which are present in that region. MEFs were also allowed to form spheroid aggregates by hanging drop method (Fig. 3F) similar to embryoid bodies (EBs) from ESCs as an important criterion to demonstrate their enhanced plasticity. Although the MEF-derived aggregates were not as tightly packed and uniform as EBs, they were similar in all other aspects. These results suggest that MEFs have an inherent multipotent nature and may extend beyond the mesodermal lineage.

Figure 3.

Expression of mesenchymal markers and differentiation potential of mouse embryonic fibroblasts (MEFs) at P5 (A) reverse transcription–polymerase chain reaction (RT–PCR) analysis of MEF and MEF aggregates for mesenchymal stem cells (MSC)MSC markers: CD 29, CD 44, SCA-1, α-smooth muscle actin (SMA) and vimentin. (B) Indirect immunofluorescence of MEFs showing positive expression for Vimentin. (C) The in vitro adipogenesis as demonstrated by Oil Red-O staining at differentiation day 21 in MEFs (magnification ×10). (D) In vitro osteogenesis evidenced by mineralized matrix deposition stained with Von Kossa staining in MEFs 3 weeks after induction. (E–F) Phase contrast micrographs showing spontaneous differentiation of MEFs into neural progenitor-like cells and aggregate formation (×10 and ×20 magnification).

Pluripotent characteristics of embryonic fibroblasts

We studied the MEFs in monolayers and aggregates/spheroids for the presence of embryonic and trilineage markers at the mRNA and protein levels. The gene expression profile of MEFs showed an upregulation of the pluripotent markers such as Oct-4, Nanog, Sox-2 and ABCG2 as compared to MEF aggregates (Fig. 4A). The mESC was used as a positive control (Fig. 4A). Further, the genotypic studies of MEFs revealed the expression of early ectoderm, mesoderm and endoderm progenitor markers; Nestin, β-III tubulin and Otx-2; MEF-2, Mesp2 and GATA-2; GATA-4, α-amylase and PDX-1 were also highly expressed in MEFs as compared to the MEF aggregates (Fig. 4B). The MEFs even demonstrated positive immunostaining for SSEA-1, Ki-67, Nestin, α-SMA, PDX-1 and C-peptide by immunocytochemistry (Fig. 4C–H). SSEA-1 represents a pluripotent marker, Ki-67 represents a cell proliferation marker and Nestin, α-SMA, PDX-1 and C-peptide represent the ectoderm, mesoderm and endoderm germ layers.

Figure 4.

(A–B) Reverse transcription–polymerase chain reaction (RT–PCR) based gene profiling of mouse embryonic fibroblasts (MEFs) in monolayers and MEF aggregates for pluripotency markers: Oct-4, Nanog, Sox-2 and ABCG2 and trilineage markers: Nestin, β-III tubulin, Otx-2 (ectoderm), MEF-2, Mesp2, GATA-2 (mesoderm) GATA-4, α-amylase, PDX-1 (endoderm). (C–H) Immunofluorescence analysis of MEFs showing expression of pluripotent (SSEA-1), cell proliferation (Ki-67) and tri-lineage (ectoderm: nestin, mesoderm: α-smooth muscle actin (SMA) and endoderm: PDX-1 and C-peptide) markers in MEFs (×10 magnification). Differentiation studies were performed with cells from P5.

Conformation of pluripotency by in vivo teratoma formation

The pluripotency of MEFs was verified by in vivo teratoma formation in Balb/c mice. Five Balb/c mice were transplanted subcutaneously on the left side with agarose plug alone and right side with agarose plug with MEF cells (Fig. 5A). After 2 weeks the animals were killed and the visible tumors underwent histological examination (Fig. 5B,C). The weight and the volume of the MEF induced teratoma with the agarose plug were significantly higher than the agarose plug alone (Fig. 5C). Histological examination of teratomas revealed the in vivo differentiation capacity of the MEFs into all the three germ layers including skin, cartilage, adipose tissue and ductal tissue (Fig. 5E–H). These results of this experiment were comparable to teratoma formation assay from mouse ES cells as standard positive control (Fig. S1).

Figure 5.

(A–C) Mouse embryonic fibroblast (MEF) (P5)-mediated teratomas in Balb/c mice on right side of mice, while saline showed no outgrowth on left side of body. (D) Graph shows comparative analysis of weight (image_n/dgd12043-gra-0001.png) and volume (image_n/dgd12043-gra-0002.png) of MEF induced teratoma and agarose plug alone after 15 days. Results are expressed as Mean ± SEM with n = 5 animals. **P-values are < 0.05. Histological images of teratoma after H and E staining confirm the representatives of three germinal layers: (E) Ectoderm (skin tissue), (F) mesoderm (cartilage), (G) Mesoderm (adipose tissue) and (H) Endoderm (ductal tissue) (×10 magnification).


The concept of ‘cell plasticity’ has attracted extraordinary interest in the past decade. It is considered to be a compensatory mechanism by which developing or regenerating tissues adjust their cell number. It may occur by cell fusion or environmental reprogramming that are mimicked in the laboratory by transfer of nuclei or introduction of transcription factors (Bonfanti et al. 2012). There is enough evidence indicating the presence of multipotent MSCs in the bone marrow, and also in prenatal, neonatal and adult ‘mesenchymal’ tissues (Bernardo et al. 2009; Hematti 2012). MSCs are defined by their ‘fibroblast-like’ morphology, characteristic plastic-adherence, specific marker expression profiles and differentiation into fat, bone and muscle (Dominici et al. 2006). Fibroblasts, on the other hand exist in virtually every organ in the human body; defined as adherent cells, which are not endothelium, epithelium or hematopoietic in origin and which have the capacity to synthesize and remodel the extracellular matrix (Haniffa et al. 2009). Interestingly several studies show that cells considered to be fibroblasts not only express surface markers almost identical to MSC, but could also differentiate into bone, fat and cartilage (Sabatini et al. 2005; Lysy et al. 2007; Covas et al. 2008; Lorenz et al. 2008; Blasi et al. 2011).

In the present study, we isolated MEFs from time pregnant mice (12.5–13.5 dpc) as a suitable model to study the plasticity of fibroblasts. Primary MEFs are believed to serve as a rich source of mesenchymal progenitor cells, and are also widely used as the feeder layer for culturing ESCs (Abbondanzo et al. 1993). However, our results indicate that MEFs exhibit greater plasticity than MSCs and are hence closer to ESCs. The isolated MEFs had a population doubling time of 24 h and exhibited the highest quiescent G0/G1 phase (66%) compared to mitotic G2/M phase (28%) and the synthesis S phase (5%). The cells could be successfully maintained in culture until P12; however, after P9 the cells exhibited a flattened morphology and the doubling time increased considerably indicating senescence as confirmed by β-galactosidase staining. This is not surprising as a similar observation was witnessed in the control BMSC cultures.

Flow cytometry analysis and immunocytochemistry revealed that MEFs were positive for MSC markers like CD90, CD44, Sca-1 and Vimentin. The MEFs readily differentiated into adipocytes and osteocytes upon induction. This could be explained by a recent study, which reported that osteoblastic differentiation of MEF required BMP-2 treatment (Saeed et al. 2012). The study also confirmed that MEF contain a subpopulation of stem cells that behave in ex vivo and in vivo assays, similar but not identical to MSC derived from bone marrow (Saeed et al. 2012). However, in addition to the MSC markers the MEFs were positive for the ESC marker SSEA-1 and proliferation marker Ki-67. The MEFs could also form embryoid body (EB)-like aggregates or spheroids by hanging drop method. To further confirm the molecular signature of the MEFs and their differentiated derivatives we performed gene expression analysis by reverse-transcriptase PCR for a panel of MSC and ESC markers. The MSC markers like CD29, CD44, Sca-1 and α-SMA and ESC markers like Oct-4, Sox-2, Nanog and ABCG2 were strongly expressed by MEFs which downregulated upon aggregate formation. Nevertheless, the MEF aggregates, unlike EBs failed to show any significant upregulation of the lineage specific markers in comparison to the monolayer cultures. Although MEFs form aggregates in suspension cultures that are structurally indistinguishable from EBs; perhaps true EBs are not formed. We reason that the inductive effects resulting from signaling between cell populations in EBs, which promote complex morphogenesis in case of ESCs are absent in MEF aggregates. Nevertheless, the consistent expression of the differentiated markers in MEFs is indicative of their inherent propensity of tri-lineage differentiation.

Further, the MEFs when injected into immunocompromised Balb/c mice formed multiple subcutaneous tumors at the end of 2 weeks similar to the ESC treated group used as positive control, as opposed to no tumors in the sham mice. Histological analysis of the tumors revealed the presence of skin-like, cartilage-like and adipocyte-like and ductal tissue-like representing the ectodermal, mesodermal and endodermal lineages. These in vivo studies confirm the in vitro ESC-like nature of MEFs.

Based on these preliminary studies, in a nutshell, it can be suggested that MEFs are not terminally differentiated cell types. They do exhibit certain stem cell-like features and can self-renew and differentiate into other cell types. Although it has been shown earlier that MEFs exhibit a similar phenotype like BMSCs (Saeed et al. 2012); this is the first study reporting MEFs that also portrays similar but not identical characteristics to ESCs. Hence, MEFs can be regarded as a connecting link during the development process starting from the embryo to the fetus to the adult stage that encompasses the prenatal and postnatal stages of development. Therefore, this unique population of MEFs can be used as a powerful tool to understand and study the regenerative biology during the complicated process of mammalian development. In addition, this report will provide valuable insights about the degree of stemness of MEF, which in turn may assist in deciding the reprogramming strategy to be used for creation of induced pluripotent stem cells from fibroblasts.

However, this study is far from exhaustive. This report provides important information with respect to the pluripotent characteristics of the MEFs; which if proven beyond doubt may be extremely significant and revolutionary. Further it is crucial to determine the role of these versatile cells during the process of embryonic development. To find out whether it is the MEFs that retain a certain pluripotent signal or is it due to another contaminated cell type that helps in balancing and retaining the embryo in their natural state is critical to successful realization of the implications of this study.


The authors acknowledge the infrastructure and financial support of Manipal University, Manipal for carrying out the present work at MIRM, Bangalore. Acknowledgement is also due to Dr Vijaylakshmi Venkatesan, NIN, Hyderabad and Shagufta Parveen for help in performing immunocytochemistry of C-peptide and teratoma experiments using mouse ESC.