Murine Embryonic Stem Cells Secrete Cytokines/Growth Modulators That Enhance Cell Survival/Anti-Apoptosis and Stimulate Colony Formation of Murine Hematopoietic Progenitor Cells
Stromal cell–derived factor (SDF)-1/CXCL12, released by murine embryonic stem (ES) cells, enhances survival, chemotaxis, and hematopoietic differentiation of murine ES cells. Conditioned medium (CM) from murine ES cells growing in the presence of leukemia inhibitory factor (LIF) was generated while the ES cells were in an undifferentiated Oct-4 expressing state. ES cell–CM enhanced survival of normal murine bone marrow myeloid progenitors (CFU-GM) subjected to delayed growth factor addition in vitro and decreased apoptosis of murine bone marrow c-kit+lin− cells. ES CM contained interleukin (IL)-1α, IL-10, IL-11, macrophage-colony stimulating factor (CSF), oncostatin M, stem cell factor, vascular endothelial growth factor, as well as a number of chemokines and other proteins, some of which are known to enhance survival/anti-apoptosis of progenitors. Irradiation of ES cells enhanced release of some proteins and decreased release of others. IL-6, FGF-9, and TNF-α, not detected prior to irradiation was found after ES cells were irradiated. ES cell CM also stimulated CFU-GM colony formation. Thus, undifferentiated murine ES cells growing in the presence of LIF produce/release a number of biologically active interleukins, CSFs, chemokines, and other growth modulatory proteins, results which may be of physiological and/or practical significance.
Murine embryonic stem (ES) cells are an in vitro copy of an in vivo population of cells known as inner cell mass [1, 2]. ES cells are pluripotent and can give rise to all cell types of the body . They can colonize germ lines resulting in chimeric animals, undergo multilineage differentiation in vitro and produce a range of well-differentiated progenitors [4, 5], suggesting a potential for ES cells in cell replacement and gene therapy. Murine ES cells are dependent for their maintenance on cytokines either secreted by themselves or in the medium that activate intracellular signals through cell surface receptors. For example, exogenously supplied leukemia inhibitor factor (LIF) is critical for maintaining ES cells in an undifferentiation stage in vitro [6–8]. Little is reported of the production/release of biologically active proteins by ES cells. We recently reported that stromal cell–derived factor-1 (SDF-1/CXCL12) secreted by and presented to murine ES cells influence their survival, migration, and production of progenitor cells . To evaluate whether other cytokines were produced or released from murine ES cells, we conditioned medium with these cells in the presence of LIF to maintain them in an undifferentiated state, and assayed the CM for cytokines, chemokines, and other growth modulatory factors by protein detection and functional assessment. We found that murine ES cells release factors that enhance survival of normal bone marrow myeloid progenitor cells and stimulate the growth of these cells. This functional activity of ES cell conditioned medium (CM) appears attributable, at least in part, to combinations of low levels of known regulatory proteins detected by immunoassay in the ES cell CM.
Materials and Methods
Cell Culture and Condition Media
E14, CCE, and RI are ES cell lines derived from 0129 mouse embryos [10, 11]. ES cell lines were cultured on gelatinized plates in Dulbecco's modified Eagle's medium (DMEM) with 15% ES cell–qualified fetal calf serum (FCS; Hyclone, Logan, UT, http://www.hyclone.com), 5.5 × 10−2 mM β-mercaptoethanol, (Gibco BRL, Carlsbad, CA, http://www.gibcobrl.com), and 103 U/ml LIF (Chemicon, Temecula, CA, http://www.chemicon.com). E14 ES cells were stained with anti-mouse Oct-4 antibody (Chemicon) to confirm morphological assessment that the cells were undifferentiated. Cells were plated at 106 cells/100-mm dish; media were changed after 16 hours. The conditioned medium (CM) in the presence or absence of serum was collected 48 hours later and stored at −80°C. After collection of CM, E14 ES cells were also stained with anti-mouse Oct-4 antibody (Chemicon) to make sure the cells were still undifferentiated (Fig. 1A, 1B). All ES cells expressed Oct-4 prior to and at the end of conditioning the culture medium in the presence of LIF and fetal bovine serum (Hyclone). In some cases, after cells reached log-phase growth, they were trypsinized, washed, and resuspended in media. Cells were irradiated at 3,000 rads with a gamma irradiation and plated at 106 cells/100-mm dish, with medium changed after 16 hours. This was done to determine potential effects of decreased proliferation of ES cells on release of factors into conditioned medium. Forty-eight hours later, CM was collected and frozen at −80°C.
Cytokine Analysis of CM
CMs were sent to Charles River Laboratories (Austin, TX, http://www.criver.com) for cytokine analysis. Cytokines and other factors capable of being detected by this analysis are shown in Table 1.
RNA Protection Assay
To confirm results of cytokine analysis, we used RNA protection assay (RPA) for assessment of mRNA expression of selected cytokines. RNA of ES cells was obtained by RNase easy kit (Qiagen, Valencia, CA, http://www1.qiagen.com). Multiprobe template set mCK-5c (BD Bioscience, San Diego, CA, http://www.bdbiosciences.com) was selected to verify the macrophage inflammatory protein (MIP)-1β, MIP-1α, MIP-2, and (inducible protein) IP-10 gene. Following instructions of multi-probe RNase protection assay system (BD Bioscience), RNA was labeled by [α-32P]UTP and hybridized with ES cell RNA. After digesting nonhybridized RNA, the sample was run on 4.75% acrylamide gel and the gel exposed overnight at −80°C.
Survival Assay for Hematopoietic Progenitor Cells in CM
Growth factors are required for hematopoietic progenitor cell (HPC) survival. Withdrawal of growth factors initiates a process of cell death. To determine the effect of CM produced by murine ES cells on survival of HPC, bone marrow (BM) cells from C57/BL6 mice were cultured in 1% methylcellulose-based Iscove's modified Dulbecco's media (IMDM) or DMEM with 30% fetal bovine serum (FBS; Hyclone), 5.5 × 10−2 mM β-mercaptoethanol, 2 mM L-Glutamine (Gibco BRL), and 0.1 mM Hemin (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) at 5 × 104 cells/ml  in the absence or presence of murine ES cell CM. Growth factors (erythropoietin 1 U/ml, stem cell factor 50 ng/ml, and 5% pokeweed mitogen mouse spleen conditioned medium) were added at 0, 24, 48, or 96 hours to each group and colonies scored 7 days after the addition of growth factors.
Apoptosis Assay for c-kit+lin− BM Cells
To analyze apoptosis of progenitors after growth factor withdrawal, we used a progenitor enrichment kit to deplete the lineage positive cells (StemCell Technologies, Vancouver, BC, Canada, http://www.stemcell.com) from C57/BL6 mouse bone marrow and a magnetic bead–separation method to obtain c-kit+ cells (Miltenyi Biotec, Auburn, CA). Cells were cultured in IMDM with 15% FBS serum without growth factor ± 15% CM. Cells were collected after 72-hour culture, stained with Annexin V antibody and propidium iodide (PI) (BD Bioscience, San Diego) for 15 minutes in the dark and analyzed using flow cytometry.
Influence of ES Cells CM on Stimulation of Colony Formation by Bone Marrow CFU-GM
To evaluate the stimulating effect of CM on bone marrow cells, we set up the colony assay with no added growth factors. Hemin enhanced stimulation of CFU-GM–derived colonies, so we evaluated cultures with/without Hemin. BM cells from C57/BL6 mice were cultured in 1% methylcellulose-based IMDM with 30% FBS (Hyclone), 5.5 × 10−2 mM β-mercaptoethanol, and 2 mM L-Glutamine (Gibco BRL) with/without 0.1 mM Hemin (Sigma-Aldrich) at 5 × 104 cells/ml  in the absence or presence of murine ES cell CM. Colonies were scored 7 days after incubation at 5% CO2, 5% (lowered) O2 and in a humidified chamber.
Significant differences were determined by t test comparisons from at least three experiments each.
Results and Discussion
Influence of Conditioned Media from ES Cells on Survival of Bone Marrow CFU-GM Colony Formation Using Delayed Addition of Growth Factors
To determine whether ES cell CM increases BM-cell survival, we used CM from three different ES cell lines (E14, CCE, and R1) that were prepared in the presence of serum. In addition, CM was also prepared from the E14 cell line in the absence of serum. The CM was collected after 48 hours. Experiments were set up with CM added at time 0 and with growth factors added at either 0, 24, 48, or 96 hours. CFU-GM colonies formed by bone marrow cells were counted 7 days after the addition of growth factors (Fig. 1C, 1D). When growth factors were added at the beginning of the experiment, there were no significant differences between the control and other groups. This demonstrates that the CM had no effect on cytokine-stimulated-CFU-GM–colony formation. Delay of the addition of growth factors for 24 hours resulted in a decrease of detectable colony formation to about 50%, and this was further decreased when the addition of growth factors was delayed for 4 days. CM prepared from E14, CCE, and R1 ES cells significantly enhanced survival of CFU-GM colonies. CM prepared from E14 cells in the absence of serum also enhanced survival of CFU-GM colonies. This demonstrates that ES cells in the presence of LIF release factors that can influence survival of HPCs.
Influence of E14 ES Cells CM on Apoptosis of c-kit+lin− BM Cells
Since ES cells release cytokines that increased survival of CFU-GM, we also checked if the ES cell CM would decrease apoptosis. c-kit+lin− cells are highly enriched for HPCs . We cultured c-kit+lin− BM cells ± CM for 72 hours without any other growth factors, stained the cells with Annexin V and PI and analyzed cells using flow cytometry (Fig. 1E, 1F). The ES cell CM dramatically decreased apoptosis in c-kit+lin− BM cells from 35 ± 3% to 18 ± 0.6%.
Proteins Secreted by E14 ES Cells
Since ES cell CM enhanced survival of CFU-GM and decreased apoptosis of c-kit+lin− cells, we evaluated what cytokines, chemokines, growth factors, and other factors might have been secreted by the ES cells growing in the presence of LIF and serum. The CM was tested for 59 different proteins (shown in Table 1). Proteins released by ES cells into the CM that were above the serum background levels and significantly above the lowest dose detectable by the assay are shown in Table 2. ES CM contained interleukin (IL)-10, IL-11, IL-1, macrophage-colony stimulating factor (M-CSF), oncostatin M (OSM), stem cell factor (SCF), vascular endothelial growth factor (VEGF), as well as a number of chemokines and other molecules. IL-11, IL-1, M-CSF, OSM, SCF, and VEGF are already known to have an enhancing effect on the survival/antiapoptosis of HPCs. Thus some or all of the survival enhancing effects may be due to these molecules or SDF-1/CXCL12 , either alone or in combination. We selected the MIP-1β, MIP-1α, MIP-2, and IP-10 genes for RPA assay. ES cells expressed these genes (Fig. 2), consistent with the protein assay data.
Influence of Irradiation on Release of Proteins from ES Cells
We compared irradiated ES cells and non-irradiated ES cells to see if irradiation had an effect on release of factors that influenced HPC survival. E14 ES cells were irradiated at 3,000 rads and plated overnight prior to washing and conditioning the medium. CM was collected after 48 hours. CM collected from irradiated ES cells was still able to enhance survival of CFU-GM as well as that of CM collected from nonirradiated ES cells (Fig. 3). However, results in Table 3 demonstrate that some proteins were significantly enhanced, and some significantly decreased by irradiating ES cells prior to their conditioning the culture medium. It is possible that the enhanced detection of some cytokines after irradiation of ES cells may be due to less of the released cytokines being used by the reproductively sterilized ES cells, compared with the nonirradiated ES cells still capable of division. Alternatively, increased release of cytokines may be due to an enhanced stress response.
Influence of ES Cell CM on Stimulation of Colony Formation by Bone Marrow CFU-GM
ES cell CM was assessed in the presence and absence of hemin for its capacity to stimulate or enhance colony formation of bone marrow CFU-GM. ES cell CM prepared in the presence of serum manifested colony–stimulating activity (CSA) (Fig. 4). ES cell CM prepared in the absence of serum also had CSA, although at a significantly lower (p < .01) level than that of the CM prepared in the presence of serum (data not shown). Hemin by itself can have modest CSA effects, as shown in Figure 4, and the effects of hemin and ES cell CM (prepared in the presence of serum) were additive to slightly above additive (Fig. 4). It is likely that the CSA of the ES cell CM was due to a combination of factors in the CM that are CSFs (e.g. M-CSF), as well as those that have the capacity to enhance the effects on, or induce release of CSFs and other cytokines (e.g. IL-1, IL-11) from the unseparated-bone marrow target cells, as well as the activity of costimulating molecules (e.g. SCF, OSM). The activities of these cytokines have been reviewed . M-CSF, IL-1, IL-11, SCF, and OSM were detected in the ES cell CM (Table 2), although at concentrations that may be below the capacity of each to act alone. Thus, we believe that the factors in the ES cell CM were acting in concert with other factors in the CM to manifest the CSA activity of the ES cell CM.
That murine ES cells have the capacity to release growth factors for survival and stimulation of myeloid progenitor cells while they are in an undifferentiated state, as documented by their expression of Oct-4 and the presence of LIF, opens up the question as to why these ES cells might be producing/releasing these proteins. It is possible that the ES cells themselves may require some of these proteins for their own survival and proliferation, as we have recently documented for SDF-1/CXCL12 , or perhaps they are of use for cells in the blastocyst that may require their presence for growth and/or differentiation.
Our studies did not distinguish what role LIF itself plays in the production/release of these growth modulatory proteins as we needed to keep LIF in the cultures to keep the ES cells in an undifferentiated proliferating form. It is possible that LIF is an inducer of these proteins but until we can keep the murine ES cells in an undifferentiated form without LIF, we cannot answer the question of whether or not LIF is important for inducing release of proteins from the ES cells.
In summary, our studies have demonstrated that undifferentiated ES cells growing in the presence of LIF can release proteins into the culture medium, some of which, probably acting together, have the capacity to enhance survival of normal bone marrow myeloid progenitor cells, as well as to stimulate their proliferation and differentiation. Aside from this being a finding that may or may not have relevance to the growth and differentiation of ES cells during embryogenesis, it is possible that these cells, alone, or when genetically manipulated to express other growth modulatory genes, may be able to serve after irradiation as accessory cells for enhancement of in vivo hematopoietic activity, or activity on other cell lineages that may respond to cytokines, chemokines, or other factors released by ES cells.
The authors indicate no potential conflicts of interest.