Equine umbilical cord blood contains a population of stem cells that express Oct4 and differentiate into mesodermal and endodermal cell types



Mesenchymal stem cells (MSCs) offer promise as therapeutic aids in the repair of tendon, ligament, and bone damage suffered by sport horses. The objective of the study was to identify and characterize stem-like cells from newborn foal umbilical cord blood (UCB). UCB was collected and MSC isolated using human reagents. The cells exhibit a fibroblast-like morphology and express the stem cell markers Oct4, SSEA-1, Tra1-60 and Tra1-81. Culture of the cells in tissue-specific differentiation media leads to the formation of cell types characteristic of mesodermal and endodermal origins. Chondrogenic differentiation reveals proteoglycan and glycosaminoglycan synthesis as measured histochemically and Sox9 and collagen 2A1 gene transcription. Osteocytes capable of mineral deposition, osteonectin and Runx2 transcription were evident. Hepatogenic cells formed from UCBs express albumin and cytokeratin 18. Multinucleated myofibers that express desmin were observed indicating partial differentiation into mature muscle cells. Interestingly, conventional human protocols for UCB differentiation into adipocytes were unsuccessful in foal UCB and adult horse adipose-derived MSC. These results demonstrate that equine UCB can be induced to form multiple cell types that underlie their value for regenerative medicine in injured horses. In addition, this work suggests that subtle differences exist between equine and human UCB stem cells. J. Cell. Physiol. 215: 329–336, 2008. © 2007 Wiley-Liss, Inc.

Bone marrow (BM) derived mesenchymal stem cells (MSCs) are the conventional model of choice for adult stem cell based therapeutics in humans due to their mutlilineage differentiation capabilities. Their relative ease of expansion in vitro without loss of plasticity makes MSCs an attractive repair aid for damaged or diseased heart, bone and vascular tissues [for review see Giordano et al. (2007)]. However, enthusiasm for the use of MSCs as cytotherapeutics is tempered by their age-dependent decline in absolute numbers and the invasive nature of their harvest (Stenderup et al., 2003). To counter these problems, umbilical cord blood (UCB) MSCs may represent a viable alternative. Several reports define a clonogenic population of cells from the umbilicus that differentiate into both mesenchymal and non-mesenchymal tissue derivatives (McGuckin et al., 2003; Aoki et al., 2004; Baal et al., 2004; Bonanno et al., 2004; Peled et al., 2004; Ruzicka et al., 2004; He et al., 2005; Holm et al., 2006; Martin-Rendon et al., 2007). The identity of these cells as circulating stem-like progenitors versus endothelial progenitors detached from the umbilicus remains debatable (Kogler et al., 2006).

A hierarchy in stem cell plasticity exists such that embryonic stem (ES) cells are pluripotent and adult MSCs are more limited in their differentiation capacity (Feinberg, 2007). UCB stem cells likely fall in the area between the two. The three classes of stem cells demonstrate variable stage specific embryonic antigen (SSEA) and tumor rejection antigen (Tra) surface marker protein expression patterns as well as differences in transcriptional circuitry. SSEA-3 and SSEA-4 are prevalent on the surface of human ES cells; these undifferentiated cells do not express SSEA-1 (Thomson et al., 1998; Reubinoff et al., 2000; Henderson et al., 2002). By contrast, mouse blastocyst inner cell mass cells and ES cells express SSEA-1 but not SSEA-3 or SSEA-4 (Henderson et al., 2002; Tielens et al., 2006). The keratan sulfate proteoglycan markers, Tra1-60 and Tra1-81, are localized within the extracellular matrix of human ES cells (Henderson et al., 2002; Stojkovic et al., 2004). Key to the establishment and maintenance of the undifferentiated state of ES cells are the coordinated activities of Oct4, nanog, and Sox2 (Boyer et al., 2005). This combination of surface markers and plasticity genes represent the minimal defining components of a naïve ES cell. By comparison, human BM derived MSCs are more limited in their expression of the central ES indicators likely owing to the heterogeneity of the population. SSEA4 is present on the surface of BM-MSC; the cells lack Oct4 but can be induced to form multiple lineages (Gang et al., 2007). Culture of BM-MSC in FGF2 supplemented media results in Oct4 and nanog transcription suggesting that a premature phenotype reminiscent of ES cells can be established (Battula et al., 2007). UCB stem cells are unique in that they possess an intermediate phenotype that more closely resembles ES cells. SSEA-3, SSEA-4, Tra1-60, Tra1-81, Oct4, and Nanog are present in this population (McGuckin et al., 2005; Zhao et al., 2006; Markov et al., 2007; Sun et al., 2007).

BM-MSC isolated from adult horses differentiate along the chondrogenic and osteogenic lineages comparable to their human counterparts (Fortier et al., 1998; Worster et al., 2001; Koerner et al., 2006). However, a reduced level of success exists for the formation of adipocytes from BM aspirates (Koerner et al., 2006; Vidal et al., 2006). Because human UCB stem cells exhibit a heightened degree of plasticity, we chose to identify a comparable cell entity in newborn foal cord blood as an alternative to BM-MSC. Using conventional human purification methods, culture conditions and differentiation protocols, an equine UCB cell population was discovered that possesses stem cell-like markers and multilineage differentiation capabilities. The isolation and characterization of these cells represent a first-step toward their application in cytotherapeutic repair of sport horse injuries.

Materials and Methods

UCB collection and stem cell isolation

Cord blood (n = 25) was collected from the intact umbilicus at foaling into a sterile 50 ml centrifuge tube containing ethylenediamine tetraacetic acid (EDTA) as an anti-coagulant. Blood was stored at 4°C and further processed within 12 h of collection. Samples were incubated for 20 min with RosetteSep Human Cord Blood Progenitor Enrichment Cocktail (50 µl/ml blood; Stem Cell Technologies, Seattle, WA), a commercially available product for negative selection of human UCB stem cells. An equal volume of phosphate-buffered saline (PBS) containing 2% fetal bovine serum (FBS, Invitrogen, Carlsbad, CA) was added. The cell suspension was layered atop a Ficoll-Paque Plus (Stem Cell Technologies) cushion and centrifuged at 1,200g for 20 min. The cell interface was collected and cultured.

Equine UCB and adipose-derived (AD) stem cell culture

MSCs isolated from equine adipose tissue were purchased from Sciencell Research Laboratories (San Diego, CA). Cells were cultured in Mesenchymal Stem Cell Medium (Sciencell Research Laboratories) on standard tissue plasticware, according to manufacturer's recommendations. UCB stem cells were cultured in Dulbecco's modified Eagle media (DMEM, Invitrogen) supplemented with 10% FBS and 5 µg/ml Plasmocin (Invitrogen). Culture medium was exchanged every three days. Cells were passaged at 70% confluency using 0.025% trypsin-EDTA (Invitrogen).

RNA isolation, reverse transcription (RT), and polymerase chain reaction (PCR)

Total RNA was isolated by lysis in STAT60 (Iso-Tex Diagnostics, Friendswood, TX) and ethanol precipitation. The RNA was digested with DNase (Ambion, Austin, TX) to remove genomic DNA contaminants. One microgram of total RNA was reverse transcribed (Superscript III, Invitrogen) in 20 µl reaction volume. Two microliters of first strand cDNA was amplified with gene-specific primers and AccuPrime DNA polymerase (Invitrogen). Primer sequences included glyceraldehyde 3-phosphate dehydrogenase (F-GATTCCACCCATGGCAAGTTCCATGGCAC, R-GCATCGAAGGTGGAAGAGTGGGTGTCACT), collagen 2a1 (F-CAGCTATGGAGATGACAACCTGGC, R-CGTGCAGCCATCCTTCAGGACAG), Sox9 (F-GCTCCCAGCCCCACCATGTCCG, R-CGCCTGCGCCCACACCATGAAG), osteonectin (F-CCCATCAATGGGGTGCTGGTCC, R-GTGAAAAAGATGCACGAGAATGAG), Runx2 (F-CGTGCTGCCATTCGAGGTGGTGG, R-CCTCAGAACTGGGCCCTTTTTCAG), albumin (F-AACTCTTCGTGCAACCTACGGTGA, R-AATTTCTGGCTCAGGCGAGCTACT) and cytokeratin18 (F-GGATGCCCCCAAATCTCAGGACC, R-GGGCCAGCTCAGACTCCAGGTGC). PCR products were visualized following electrophoresis through 2% agarose gels containing ethidium bromide. Representative images were captured with a Kodak ImageDoc system and inverted in Adobe Photoshop CS.

Osteogenic differentiation

Cells were plated at a density 1,300 cells/cm2 and allowed to attach overnight in normal growth medium. The following day, cells were washed twice with PBS and placed in an osteogenic differentiation medium composed of alpha modified Eagle medium (α-MEM), 10 mM β-glycerophosphate, 0.1 µM dexamethasone, 0.1 mM ascorbic acid (Tondreau et al., 2005; Wagner et al., 2005). Media was changed twice weekly. Cells were fixed in 4% paraformaldehyde in PBS for 15 min on days 7, 14, and 21. Total RNA was isolated from parallel plates.

Chondrogenic differentiation

UCB stem cells were pelleted to a micromass, promoting chondrogenic differentiation in a three-dimensional environment. Cells (4 × 105) were pelleted at 1,000g for 5 min. The medium was removed and 0.5 ml chondrogenic medium was added (Worster et al., 2000; Tondreau et al., 2005). Chondrogenic medium consisted of DMEM, 1.0 g/L insulin, 0.55 g/L transferrin, 0.67 mg/L sodium selenite (ITS-X, Invitrogen), 10 ng/ml transforming growth factor beta-1 (TGF β1, R&D Systems, Minneapolis, MN), 35 µg/ml ascorbic acid and 100 nM dexamethasone (Sigma, St. Louis, MO). Media was changed twice weekly. After 7, 14, and 21 days, the micromass was embedded and frozen in OCT freezing compound. Alternately, micromasses were washed with PBS and used for RNA isolation.

Adipogenic differentiation

UCB stem cells were plated at a density of 3,000 cells/cm2 in growth medium. Adipogenic differentiation was induced with Iscove's modified Dulbecco's media (IMDM) supplemented with 10% FBS, 1 µM dexamethasone, 10 µg/ml recombinant human insulin, 0.25 mM 3-isobutyl-1-methylxanthine (IBMX) and 100 µM indomethacin (Wagner et al., 2005). Medium was replaced every three days. Cells were fixed with 4% paraformaldehyde in PBS after 7, 14, and 21 days in culture.

Hepatogenic differentiation

UCB stem cells were plated at a density of 5,000 cells/cm2 and allowed to attach overnight in normal growth medium. The following day, cells were washed twice with PBS and placed in hepatogenic medium [1% FBS, 20 ng/ml recombinant human hepatocyte growth factor (HGF, R&D Systems), 10 ng/ml recombinant human fibroblast growth factor 4 (FGF4, R&D Systems) in IMDM], as described (Kang et al., 2005). Medium was replaced twice weekly. On days 7 and 14 cells were fixed in 4% paraformaldehyde in PBS for 15 min. Total RNA was isolated from parallel plates.

Myogenic differentiation

UCB stem cells were plated at 3,000 cell/cm2 on gelatin-coated tissue cultureware. Differentiation was initiated by incubation in low glucose DMEM supplemented with 200 µg/ml galectin-1 (R&D Systems) essentially as described (Chan et al., 2006). After 14 days, cells were fixed with 4% paraformaldehyde or lysed for RNA isolation.


Alkaline phosphatase enzymatic activity was detected colorimetrically using Nitro-Blue Tetrazolium Chloride (NBT) and 5-Bromo-4-Chloro-3'-Indolyphosphate p-Toluidine (BCIP; Pierce, Rockford, IL) following fixation with 4% paraformaldehyde. Oil Red O (0.1% in 60% isopropanol) was used to visualize lipid droplets. Alcian Blue (1% in 3% acetic acid) staining was used to detect glycosaminoglycans. Safranin O (0.1% in water) was used for the visualization of proteoglycans and cartilage. Alizarin Red (2% in water, pH 4.2) was used for the detection of mineral deposits. Calcium deposits were detected by the method of von Kossa using 1% silver nitrate and 5% sodium thiosulfate.


Cells were fixed in 4% paraformaldehyde for 10 min at room temperature and permeabilized with 0.1% Triton X-100 in PBS. Non-specific antigen sites were blocked with 5% horse serum. For the detection of stem cell markers, anti-Oct4 (1:50, Santa Cruz, Santa Cruz, CA), anti-SSEA-1, anti-SSEA-3, anti-SSEA-4 (1:50, R&D Systems), anti-Tra1-60 (1:50, Abcam, Cambridge, MA) and anti-Tra1-81 (1:50, Abcam) were used. Myogenic cells were incubated with anti-desmin (1:200, DE-U-10, Sigma Aldrich, St. Louis, MO) and Texas Red conjugated Phalloidin (Invitrogen). All antibodies were diluted in PBS containing 1% horse serum and incubated with fixed cells for 1 h at room temperature. Immune complexes were visualized with goat anti-mouse AlexaFluor488 (1:200), goat anti-rabbit AlexaFluor488 (1:200), or goat anti-rabbit AlexaFluor568 (1:200) on a Nikon T200 microscope equipped with epifluorescence. Images were captured with NIS Elements (Nikon Instruments, Melville, NY) software and compiled with Adobe PhotoShop CS.


Foal umbilical cord blood contains an Oct4-expressing cell population

Multipotent stem cells are routinely isolated from fresh cord blood at birth from humans by density gradient centrifugation. Initial Ficoll gradient separation of equine UCBs yielded a heterogeneous population with poor recovery of adherent cells. As such, a negative selection procedure (RosetteSep) was employed with the potential to remove extraneous natural killer cells, macrophages, lymphocytes, and B-cells. Significantly fewer cells were present in the buffy coat following centrifugation through Ficoll that attached readily to plastic cultureware. To ascertain their identity, cells were fixed and evaluated by immunocytochemistry for the stem cell markers, SSEA-1, Oct4, Tra1-60 and Tra1-81. Results demonstrated the presence of Oct4 in the nuclei of greater than 90% of the cells (Fig. 1). A similar percentage of the cells contained Tra1-60 and Tra1-81 as well as SSEA-1. In data not shown, alkaline phosphatase enzymatic activity was readily detected and minor amounts of SSEA-4 were noted by immunocytochemistry. Thus, newborn foal cord blood contains a cell population that can be isolated with conventional human reagents and protocols and that express characteristic ES cell marker proteins.

Figure 1.

Foal UCB cells express stem cell marker proteins. UCB stem cells (passage 5) were fixed and incubated with antibodies directed against Oct4, SSEA-1, Tra1-60 or Tra1-81. Immunoreactivity was detected with goat anti-mouse AlexaFluor 488 or anti-rabbit AlexaFluor 568 (Oct4). Nuclei were counterstained with Hoechst 33245. Scale bar = 10 µm.

UCB stem cells form chondrocytes

The plasticity of UCB stem cells was evaluated by differentiation into cartilage precursors. Adherent stem cells were pelleted and cultured as a micromass in a defined media supplemented with ascorbic acid, dexamethasone and TGF-β1. At the end of 3 weeks in chondrogenic media, the cell pellet was cryopreserved or lysed for RNA isolation. Alcian blue histology revealed the presence of proteoglycans, a matrix component of mature chondrocytes (Fig. 2A). The extracellular constituents of the micromass were rich in glycosaminoglycans, as indicated by Safranin staining. Sox9 is a transcription factor that positively regulates the expression of collagen and extracellular matrix genes in chondrocytes (Ng et al., 1997; Bi et al., 1999; Lefebvre et al., 2001). Expression levels of Sox9 and collagen 2a1 were evaluated by RT-PCR after 7 and 21 days in chondrogenic differentiation media. Gene transcripts for Sox9 are evident after 7 days but absent by 21 days in differentiation permissive media (Fig. 2B). Abundant amounts of collagen 2a1 mRNA was evident at both time frames. These results demonstrate temporal and specific activation of the chondrogenic gene program.

Figure 2.

Induction of chondrogenesis in foal UCB stem cells. Micromass cultures were established as described. Cell pellets were embedded in OCT freezing medium and 10 µm cryosections were collected. Alcian Blue and Safranin O histology indicated the presence of glycosaminoglycans and proteoglycans (A). Total RNA was isolated from cells treated in an analogous manner. RT-PCR using gene-specific primers indicated Sox9 and collagen 2A1 expression after 7 days in chondrogenic medium (B). Sox 9 mRNA was not detected at day 21. RT, reverse transcriptase.

Differentiation of UCB stem cells into osteocytes

Young thoroughbred racehorses are prone to debilitating bone fractures whose repair may be aided by stem cell-based therapeutics. Therefore, UCB stem cells were cultured on plasticware in a defined media capable of inducing osteocytes from human UCB stem cells (Tondreau et al., 2005). After 3 weeks, cells were fixed with paraformaldehyde or harvested for RNA isolation. Alizarin Red histology indicated that the putative bone cells were capable of calcium deposition (Fig. 3A). In a similar manner, Von Kossa staining detected calcium aggregates. RT-PCR confirmed the osteogenic program in these cells. Primers specific for osteonectin and Runx2 amplified products of the correct size (Fig. 3B). These results demonstrate that UCB stem cells are a source of osteocytes under appropriate in vitro cultivation conditions.

Figure 3.

UCB stem cells form osteocytes. UCB stem cells were maintained for 21 days in media containing β-glycerophosphate, dexamethasone, and ascorbic acid. Cells were fixed and stained for calcium and mineral deposition with Alizarin Red and von Kossa. Representative brightfield images (left part) and corresponding phase contrast fields (right part) at 200× are shown (A). Total RNA was isolated after 21 days in osteogenic induction medium. RT-PCR using gene-specific primers for osteonectin and RunX2 was performed. Products were separated through agarose gels and visualized with ethidium bromide (B). RT, reverse transcriptase.

Foal UCB stem cells can differentiate into endodermal-derived cell types

A key feature of ES cells is their potential to contribute to any tissue type in the body. Adult stem cells possess a more limited plasticity than their embryonic counterparts. The ability of foal UCB stem cells to differentiate into hepatocytes, a cell type that originates from the endoderm, was examined. In brief, UCB stem cells were cultured for 2 weeks in media that supports hepatocyte formation in human UCB stem cells (Kang et al., 2005). Subsequently, cells were fixed with paraformaldehyde or lysed for RNA isolation. As shown in Figure 4A, a change from an elongated, spindle-shaped morphology to one exhibiting a larger cytoplasmic volume with an elliptical shape occurs in response to the treatment media. These cells express mRNA for both albumin and cytokeratin 18, definitive markers of hepatoctyes (Fig. 4B). Equine UCBs maintained in the absence of induction media failed to express the liver marker genes (data not shown). The ability to respond in a manner similar to ES cells and form hepatocytes suggests that our UCB cell population may be more plastic than other adult MSC.

Figure 4.

Foal UCB stem cells form hepatocytes. UCB stem cells were cultured for 14 days in absence or presence of HGF and FGF4. Cells were fixed with 4% paraformaldehyde and representative phase-contrast images were captured at 200× (A). Separate cultures were analyzed by RT-PCR for cytokeratin 18 (CK18), albumin and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene expression (B). RT, reverse transcriptase.

Inefficient formation of myocytes and adipocytes by UCB cells

Koerner et al. (2006) reported limited formation of adipocytes from adult horse BM-derived MSCs. Thus, we compared the adipogenic differentiation capabilities of foal UCB stem cells and adult horse adipocyte-derived MSCs (AD-MSC). In brief, both cell types were incubated for 21 days in adipocyte induction media. Cells were fixed and evaluated by Oil Red O histology for the presence of lipid droplets. In our hands, neither UCB nor AD-MSC efficiently formed adipocytes. Sporadic fat cells containing limited amounts of lipid droplets were evident in foal UCB cell cultures; no Oil Red O positive cells were found in the AD-MSCs. This restricted differentiation profile by the two forms of stem cell was further exemplified following their incubation in myocyte induction media. UCB stem cells and AD-MSCs were incubated for 7 days in media supplemented with galectin-1, a glycoprotein that promotes myogenesis in human fetal MSCs (Chan et al., 2006). Subsequently, cells were fixed and immunostained for the skeletal muscle marker protein, desmin. UCB stem cell cultures contained several multinucleated, spindle-shaped cells that are reminiscent of myocytes. Anti-desmin immunofluorescent detection reveals that these structures express the intermediate filament protein (Fig. 5B). No desmin expressing muscle cells were present in the AD-MCSs treated in a similar manner (data not shown). The presence of organized actin filaments was examined using Texas Red conjugated phalloidin. Equine UCB-derived myoblasts contained organized actin structures throughout their cytoplasm (Fig. 5C). By contrast, AD-MSC cells contained fewer phalloidin-reactive filaments. The cytoskeletal structures pointed to distinct differences in overall cellular morphology between the differentiated AD-MSC and UCB myoblasts. UCB myoblasts were thin, elongated and cylindrical in shape whereas the AD-MSC cells were fibroblast-like with an enlarged cytoplasmic space. Our results demonstrate differences between the two types of stem cells and suggest that foal UCB cells are more plastic than adult horse MSCs.

Figure 5.

Incomplete initiation of adipogenesis and myogenesis in foal UCB stem cells and AD-MSCs. Foal UCB and AD-MSC cells were cultured in adipogenic induction media for 21 days prior to fixation. Oil Red O histology demonstrates very few adipocytes in foal UCB cells (A). UCB stem cells were cultured for 7 days in myogenic induction media prior to fixation and desmin immunostaining. Arrow indicates multinucleated cell (B). Actin filaments were detected by incubation with Texas Red-phalloidin (C). Total nuclei were visualized with Hoechst 33245. Scale bar = 25 µm.

AD-MSC do not express the same complement of stem cell markers

The inability of AD-MSC to form adipocytes was surprising given that they originate from the fat depot. To ensure that the cells were naïve and undifferentiated, subconfluent cultures of AD-MSC were immunostained for stem cell markers. Similar to UCB stem cells, AD-MSCs express Oct4, Tra1-60 and Tra1-81 (Fig. 6). However, SSEA-1 and SSEA-4 were undetectable. These results indicate that AD-MSC retain markers of adult stem cells but do not express those more closely associated with ES cells.

Figure 6.

AD-MSC fail to express embryonic stem cell markers. AD-MSCs were fixed and incubated with antibodies directed against Oct4, SSEA-1, SSEA-4, Tra1-60 or Tra1-81. Immunoreactivity was detected with goat anti-mouse AlexaFluor 488 or anti-rabbit AlexaFluor 568 (Oct4). Nuclei were visualized with Hoechst 33245. SSEA-1 and SSEA-4 were not detected. Scale bar = 10 µm.


There is widespread interest in tendon, ligament and cartilage repair in horses through the use of directed stem cell transplantation methods. To date, published reports of multipotent cells isolated from BM, peripheral blood, and UCB exist (Fortier et al., 1998; Saito et al., 2002; Koerner et al., 2006; Li et al., 2006). Cells from each of these sources display limited differentiation into mesodermal cell types with predominant induction of chondrogenic and osteogenic precursors. Beyond these two cell types, vast differences in differentiation efficiencies and alternate cellular identities exist. The disparities may be attributed to tissue source or suboptimal culture conditions; both possibilities necessitate further study. Alternatively, the transcriptional regulators that govern pluripotency may be absent or inactive thereby, limiting plasticity. Key to the ES cell-like nature is expression of Oct4, Sox2, nanog, c-myc, and Klf4 (Takahashi and Yamanaka, 2006). Foal UCB stem cells maintained in a growth factor rich medium expressed Oct4, SSEA-1, Tra1-60 and Tra1-81, all stem cell marker proteins. However, repeated attempts to detect nanog and Sox2 mRNA were unsuccessful. The absence of these transcription factors may contribute to the restricted types of cells generated and their incomplete differentiation (myoblasts). Interestingly, the ability of UCB stem cells to express these embryonic markers sets them apart from adult MSCs. Surface expression of SSEA-1 and SSEA-4 were not evident in equine AD-MSCs. The lack of SSEA markers points to a hierarchy in plasticity that may account for some of the differences in differentiation capabilities. Efforts to define culture medias that support nanog, Sox2, and Klf4 expression may lead to an increased range of differentiated lineages from UCB stem cells.

Stem cells isolated from the umbilical cord matrix of pigs develop a morphology that resembles that reported by others for equine UCB stem cells (Carlin et al., 2006; Koch et al., 2007). In both examples, the majority of the cells attached to the cultureware surface and possessed a flat, spindle-shaped, fibroblast-like morphology. A lesser population formed light-refractile colonies that grew upward from the substratum surface in a manner consistent with transformed fibroblast foci. These colonies of small cells with a high nuclear to cytoplasmic volume were evident in our cultures of newborn foal UCB stem cells only after reaching confluency. Our UCB stem cells were maintained as a monolayer and passaged at approximately 60% confluency thereby, selecting against the development of these cell clusters that appear to grow independent of contact inhibition. While the identity of this cell population remains less clear, it is possible that these colonies represent a more primitive progenitor cell. Indeed, these cell clusters resemble those found in cultures of mouse ES cells. As such, one would predict that confluent equine UCB cultures that contain both the fibroblast-like and light-refractile cell colonies would express the plasticity genes, nanog and Sox2. However, expression of SSEA-1, Tra1-60, Tra1-81 and alkaline phosphatase, in a manner consistent with equine inner cell mass-derived ES cells, provides encouraging evidence that our monolayer cells are naive and undifferentiated (Takahashi and Yamanaka, 2006).

Equine UCB stem cells, in our hands, are not direct equivalents to human UCB stem cells but do possess many similarities. Human UCB stem cells can be isolated directly from the blood and frozen without expansion (Lee et al., 2005). This aspect of enrichment and storage remains elusive in our equine UCB cells. Partial purification by negative immunoselection and density gradient centrifugation produces a cell population that survives immediate cryopreservation very poorly. This may be due to the small numbers of stem cells and/or heightened sensitivity of these cells to plasma membrane perturbation. Culture of the fresh isolates for 3–5 days allows for the removal of contaminating lymphocytes and cellular debris and expansion of the putative stem cell population, which can be stored in liquid nitrogen and subsequently recovered. Direct enrichment of the UCB stem cell population by affinity purification with CD133 antibodies may provide an alternative to both cell heterogeneity and cryopreservation issues.

The capacity of foal UCB stem cells to initiate hepatocyte-specific gene transcription demonstrates an endodermal developmental potential. Reports exist demonstrating hepatocyte formation from human UCB stem cells and MSC isolates from BM (Hong et al., 2005; Talens-Visconti et al., 2006). However, this is the first report of hepatocyte formation using equine multipotential cells. Putative stem cells from the inner cell mass of equine blastocysts undergo spontaneous differentiation in vitro to yield cell derivatives of the three germ layers with endoderm defined by RT-PCR detection of α-fetoprotein (Li et al., 2006). The ability of newborn foal UCB stem cells to form liver cells is encouraging as it provides additional evidence for a population with plasticity characteristics that more closely resemble an ES cell than an adult stem cell. Additional endoderm-derived cell types of clinical importance include pancreatic and cardiogenic. Human UCB stem cells can be induced to form heart cells following a two-step differentiation protocol that involves 5-azacytidine treatment (Kadivar et al., 2006). Culture with the hypomethylating agent suggests that UCB stem cells are more restricted in their differentiation capabilities than ES cells and require chemical-induced reprogramming. In our hands, treatment of foal UCB stem cells with 5-azacytidine did not induce the expression of myosin immunopositive cells. Because our antibody (MF20) recognizes all forms of sarcomeric myosin, this result provides indirect evidence that a full-fledged cardiocyte is not created in response to epigenetic modification. However, a more comprehensive analysis of growth factor, morphogen and substratum requirements for UCB stem cell differentiation into cardiocytes is warranted.

Induction of the myogenic gene program has proven difficult in MSC originating from multiple animal and tissue sources. Exposure of rat BM-derived MSCs to 5-azacytidine caused differentiation into elongated, multinucleated myofibers (Wakitani et al., 1995). However, reprogramming the equine UCB transcriptome with this chemical did not induce the myogenic gene program (data not shown). A similar result was noted by others (Chan et al., 2006). Others reported that human UCB stem cells formed limited numbers of desmin immunopositive cells following in vitro differentiation (Nunes et al., 2007). These cells were devoid of the myogenic regulatory factors (MRFs) as measured by RT-PCR. Interestingly, injection of the putative stem cells into mdx mice resulted in engraftment suggesting that components within the muscle niche are essential for myogenic progression. One of those proteins is likely galectin-1, a glycoprotein of the basal lamina. Chan et al. (2006) demonstrated that culture of human fetal MSC in media containing galectin-1 initiated both biochemical and morphological differentiation into myocytes. These cells formed large, multinucleated fibers that expressed contractile proteins and the MRFs. We used a similar approach with some degree of success. Supplementation of foal UCB cell culture medium with purified galectin-1 caused myogenic lineage establishment as determined by desmin immunocytochemistry. However, a large percentage of the myoblasts were fusion-defective. In addition, we were unable to detect gene transcripts for MyoD, an early MRF, or myogenin, an MRF required for fusion and contractile gene expression. In accordance with our failure to amplify members of the MRFs, we did not detect myosin heavy chain or troponin T by immunocytochemical methods. The constraints to full activation of the myogenic program may be attributed to the absence of complementary soluble proteins. The source of galectin-1 used by Chan was spent media from COS cells that produce and secrete the glycoprotein. Thus, additional proteins within the galectin-1 supplement may have aided induction of myogenesis. Alternatively, specie-specific differences may underlie the discrepant results.

Given the relative ease of adipocyte formation by human and rodent MSC, the inefficiency of adipogenesis in equine UCB stem cells was surprising. Less than 1% of cells contained Oil Red O reactive lipid droplets following application of conventional adipogenic induction protocols. Koerner reported a similar result using BM-derived and peripheral blood-derived MSC isolated from adult horses (Koerner et al., 2006). A very small number of adipocytes were found and the cytoplasmic lipid droplets within said cells were miniscule. By contrast, robust lipid formation is evident by Oil Red O histology in equine UBCs cultured in a similar adipocyte induction media (Koch et al., 2007). The discrepancy between these various reports may be attributed to the heterogeneity of the starting population and/or culture conditions. Koch reported the presence of dome-like, clusters of small cells as well as a fibroblast-like cell type (Koch et al., 2007). While we observe the same morphologies, care was taken to maintain the adherent monolayer exclusive of the foci-like colonies. Future efforts will concentrate on resolving the identity of these divergent cellular phenotypes and their contribution to plasticity.


The authors thank Cloverleaf Farms and Lambholm South for collection of newborn foal UCB. This work was supported by a grant from the Florida Pari-Mutuel Racing Trust Fund to SEJ. SAR is a Fellow of the United States Department of Agriculture National Needs grants program.