Human marrow stromal cell (hMSCs) were recently shown to expand rapidly in culture when plated at a low density of approximately 3 cells/cm2. Low-density plating promoted proliferation of small recycling stem (RS) cells that appeared to be the most multipotent cells in the cultures. Here we demonstrated that MSCs from rat bone marrow (rMSCs) are even more sensitive to low-density plating than hMSCs. When plated at approximately 2 cells/cm2, the cells expanded over 4,000-fold in 12 days, over twice the maximal rate observed with hMSCs. Analysis by fluorescence-activated cell sorter demonstrated that rMSCs had the same heterogeneity seen with hMSCs in that the cultures contained both small rapidly RS cells and much larger mature cells (mMSCs). The rat mMSCs differed from human mMSCs in that they regenerated RS cells in culture. Also, after low-density plating, colonies of rMSCs expanded into confluent cultures, whereas colonies of hMSCs did not.
Marrow stromal cells (MSCs), also known as mesenchymal stem cells or colony-forming units (CFU) fibroblastic, are nonhematopoietic multipotent stem-like cells that adhere to culture dishes. They are capable of clonal expansion in culture, support hematopoietic stem cell proliferation, and demonstrate extensive differentiation capacity . MSCs share characteristics with other multipotent stem cells such as neural stem cells, hematopoietic stem cells, side population cells, and liver stem cells because they possess the ability to self-renew and give rise to differentiated progeny [1-14].
MSCs were first described in the 1970s by Friedenstein et al., who discovered that the cells adhered to tissue culture plates, resembled fibroblasts in vitro, and formed colonies . He and others demonstrated MSCs could differentiate into bone, cartilage, and adipocytes [13, 16, 17]. These characteristics have been identified in MSCs from numerous species including humans, rats, mice, and monkeys [18-22].
In addition to bone, cartilage, and fat, MSCs have demonstrated the potential to differentiate down the myogenic pathway. Wakitani et al. demonstrated MSC differentiation to muscle in vitro . Ferrari et al. provided evidence that MSCs underwent myogenic differentiation in areas of induced muscle degeneration after infusion into immunodeficient mice . Recently, there has been increasing evidence that MSCs are capable of differentiating into neurons and astrocytes in vitro and in vivo [11, 25-28]. The results suggest that MSCs are capable of differentiating into both mesenchymal and nonmesenchymal lineages.
Here we analyzed MSCs from rats to characterize their growth properties in vitro and to establish the similarities and differences in vitro between human MSCs (hMSCs) and rat MSCs (rMSCs).
Materials and Methods
Cell Harvest and Culture
rMSCs were harvested from the bone marrow of the femurs and tibias of 6- to 12-month-old Lewis rats (Harlan Laboratories; Indianapolis, IN) by inserting a 21-gauge needle into the shaft of the bone and flushing it with 30 ml of complete α-modified Eagle's medium (αMEM) containing 20% fetal bovine serum (FBS) (lot selected for promoting rapid expansion of MSCs; Atlanta Biologicals; Norcross, GA), 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 25 ng/ml Amphotericin B. Cells were filtered through a 70-μm nylon filter (Falcon; Franklin Lakes, NJ) and the cells from one rat were plated into one 75 cm2 flask. The cells were grown in complete αMEM at 37°C and 5% CO2 for 3 days, the medium was replaced with fresh medium, and the adherent cells were grown to 90% confluency to obtain samples here defined as passage zero (P0) cells. hMSCs were harvested and cultured as described by Colter et al. . All animal work was performed under guidelines determined by institutional committees for animal welfare and use of human subjects.
Passing MSCs and Colony-Forming Assays
rMSCs at P0 were washed with phosphate-buffered saline (PBS) and detached by incubation with 0.25% trypsin and 0.1% EDTA (Cellgrow; Herndon, VA) for 5 to 10 minutes at 37°C. Complete medium was added to inactivate the trypsin. The cells were centrifuged at 450 × g for 10 minutes, the medium was removed, and the cells were resuspended in 1 to 10 ml of complete medium. The cells were counted in duplicate using a hemacytometer and then plated as P1 in 58 cm2 plates (Becton Dickinson; Falcon, Franklin Lakes, NJ) at densities ranging from 0.5 cells/cm2 (low-density) to 5,000 cells/cm2 (high-density). Complete medium was replaced (refeeding) every 3 to 4 days over the 12- to 14-day period. For the CFU assay, cells were grown for 12-14 days, culture dishes were stained with 3% crystal violet solution in 100% methanol for 10 minutes, and colonies were counted. All cells used for the experiments were P5 or earlier.
Cells from three culture plates at each plating density were trypsinized, centrifuged, resuspended, and counted. Propidium iodide (PI) (5 μg/ml) was added for 5-10 minutes to the MSCs and PI+ cells were gated out prior to phenotypic analysis. MSCs were analyzed for size, granularity, epitope expression, and green fluorescent protein (GFP) expression using a cell sorter (FACS Sort; Becton Dickinson).
Approximately 200,000 rMSCs and hMSCs were centrifuged at 450 × g for 10 minutes. The medium was removed and the pellet was resuspended in 1 ml of 100% methanol at 4°C for 10 minutes to fix the cells. The pellet was washed in 3 ml of PBS and resuspended in 1 ml of PBS containing 1% bovine serum albumin (BSA) and 0.1% serum for 1 hour at room temperature. The cells were washed in PBS, centrifuged, and resuspended in 0.5 ml of PBS containing primary antibody (1:100 dilution for a final concentration of 10 to 20 μg/ml) for 40 minutes at room temperature. For the isotype control, nonspecific mouse IgG was substituted for primary antibody. The cells were washed and resuspended in 0.5 ml PBS containing a 1:500 dilution of secondary antibody (biotin-conjugated anti-mouse IgG; DAKO; Santa Barbara County, CA) for 20 minutes at room temperature. The cells were washed, centrifuged, and then incubated in 0.5 ml PBS containing a 1:500 dilution of streptavidin-fluorescein isothiocyanate at room temperature for 20 minutes. The cells were washed a final time, resuspended in 0.5 ml PBS, and then analyzed by FACS. Antibodies for CD4, CD11b, CD43, CD45, CD59, CD90, and mononuclear phagocyte marker were from PharMingen (San Diego, CA). Antibodies to CD31 were from Chemicon (Temecula, CA).
Transduction of rMSCs to Express GFP
To investigate interconvertability of RS cells and mature MSCs, rMSCs were transduced to express GFP. As a first step, the plasmid LXSN-GFP (LXSN; Clontech; Palo Alto, CA) was made by ligating an 800 bp BamHI fragment containing the gene for enhanced GFP (EGFP-1; Clontech) into the BamHI site of a Moloney murine leukemia virus-derived plasmid .
As a next step, Phoenix amphotropic packaging cells were obtained from the American Type Culture Collection (Rockville, MD) with permission from Dr. G. Nolan (Stanford University) and they were transfected with LXSN-GFP by calcium phosphate precipitation [31, 32]. Briefly, 24 hours prior to transfection, 2.5 × 106 Phoenix cells were plated in 9.6 cm2 plates in growth medium (GM) (10% heat-inactivated FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine in Dulbecco's modified Eagle's medium [DMEM]). Just prior to transfection, the medium was changed to GM containing 25 μM chloroquine. The transfection cocktail was prepared by adding 500 μl 2× buffered saline solution (50 mM HEPES, pH 7.05; 10 mM KCl; 12 mM Dextrose; 280 mM NaCl; 1.5 mM Na2HPO4) to an equal volume of transfection mixture containing 10 mg plasmid DNA and a final concentration of 250 mM CaCl2. The transfection cocktail was added to the Phoenix cells, the cells were incubated at 37°C for 10 hours, and the medium was changed to fresh GM without chloroquine. Viral supernatants were collected after 48 hours, filtered through a 0.45 μm filter (Millipore), and stored at −80°C for further use.
As a final step, rMSCs were transduced as previously described . Briefly, 100,000 rMSCs were plated the day before infection in 9.6 cm2 plates. At the time of infection, 2.5 ml complete medium containing 20% heat-inactivated FBS, 500 μl viral supernatant and 8 μg/ml polybrene (Sigma; St. Louis, MO) were added to the cells. The infection procedure was repeated after 24 hours. After 72 hours, the medium was replaced with fresh complete medium containing 20% FBS (not heat-inactivated). Forty-eight hours after the final infection, cells were split 1:2 in 58 cm2 plates and incubated in complete medium containing 200 μg/ml G418 (Sigma) for a period of 14 days. The surviving cells were pooled and expanded for experiments.
Osteogenic Differentiation of rMSCs
rMSCs transduced with LXSN-GFP were grown to approximately 80% confluency and then transferred to osteogenic medium containing αMEM, 20% FBS, 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 25 ng/ml Amphotericin B, 10–8 M dexamethasone, 0.2 mM ascorbic acid 2-phosphate, and 10 mM beta glycerol phosphate. The osteogenic medium was replaced every 3-4 days. Mineralization was assessed after 2-3 weeks by staining with 40 mM Alizarin red (pH 4.1; Sigma) .
Effect of Plating Density on rMSCs
rMSCs (Fig. 1) were plated at densities of 2 and 16 cells/cm2 in 58 cm2 plates. After 12 days, total yield per cell plated was approximately 300 for cells plated at 16 cells/cm2, but was approximately 800 for cells plated at 2 cells/cm2 (Fig. 2). Therefore, the cells were similar to hMSCs in that the yields increased at low-plating densities . In contrast to hMSCs, the yields were significantly increased by frequently replacing the medium. Replacing with fresh medium every 3-4 days increased yield per cell plated at 2 cells/cm2 to about 4,000 cells (Fig. 3A and 3B, Figure 3.). rMSCs also expanded about twofold greater in αMEM than in DMEM (not shown).
Values for percent CFU for rMSCs were dramatically affected by small changes in plating density (Fig. 4A). In contrast, even though hMSCs required low-density plating for rapid expansion , the observed values for percent CFU were not affected by plating at low densities that varied from 1.0 to 8 cells/cm2 (Fig. 4B). This indicated that rMSCs are more sensitive to plating density than hMSCs.
FACS Analysis of rMSCs
FACS analysis for size and granularity demonstrated that cultures of rMSCs contain the same three subpopulations of cells present in cultures of hMSCs : A) large and apparently mature cells (mature MSCs); B) small agranular cells referred to as recycling stem cells (RS-1 cells), and C) small granular cells (RS-2 cells). As with hMSCs , stationary cultures of rMSCs contained only mature MSCs and RS-1 cells (Fig. 5A). RS-2 cells appeared 3-4 days after low-density plating (Fig. 5B) and then decreased in proportion as the mature MSCs expanded (Fig. 5C). In contrast to hMSCs, there was no correlation between the percent RS cells in cultures and assayed values for CFUs for rMSCs (not shown).
Ability of Mature MSCs to Reform RS Cells and Colonies
To examine the interconvertability among the three subpopulations of cells, rMSCs were transduced with LXSN-GFP. Stably transduced cells were grown in culture for 3 weeks and then analyzed by FACS (Fig. 6A and 6B, Figure 6.). The highest GFP-expressing cells were the mature rMSCs. These cells were sorted and then plated at low density to determine proliferation and colony-forming capabilities. The sorted mature rMSCs reformed RS cells (Fig. 6C). The percent of CFU obtained from presorted cultures (34%) and post-sorted mature MSCs (32%) was almost identical (not shown). The sorted mature rMSCs were also capable of osteogenic differentiation (Fig. 7). Therefore, the mature rMSC subpopulation in rat cultures contained cells that were capable of rapid expansion, colony formation, and osteogenic differentiation.
Epitope Analysis of rMSCs and hMSCs
To establish the amount of hematopoietic contamination within the MSC cultures and compare rMSCs and hMSCs, epitope analysis for hematopoietic markers was performed on rMSCs and hMSCs (Table 1). Most markers for hematopoietic cells were completely negative for both rMSCs and hMSCs including CD4, CD11b, CD43, and CD45. CD31, a marker for platelet endothelial cell adhesion molecule-1, was positive on all mature hMSCs, positive on most human RS cells, and negative on all rMSCs. CD59, an SCA-1 homologue, was positive on most of the RS and mature cells in the human cultures, but was positive only on the mature rMSCs. CD90 (thy1.1) was positive on most cells from both subpopulations for both species.
Table Table 1.. Epitope analysis for hematopoietic contamination of rMSCs and hMSCs. Cultures were analyzed and compared for the two main subpopulations: RS-1 and mature MSCs.
Mature MSC population
All cells were negative for most markers analyzed. CD90 was positive on most MSCs and CD59 was positive on all cells except the RS subpopulation of rMSC.
aMost cells are positive.
bAll cells are positive.
cCD90 is also known as Thy 1.1.
dCD59 is an SCA-1 homologue.
eCD31 is an antibody to PECAM-1.
fCD4 is an antibody to T cells.
gCD11b is an antibody to MAC-1.
hCD43 is an antibody to leukosialin (an adhesion molecule for leukocytes that invade an area during an inflammatory response).
iCD45 is an antibody to leukocyte common antigen.
jMP (mononuclear phagocyte marker) is an antibody to macrophages.
When rMSCs were plated at low-density and cultured for 4 weeks, the colonies began to expand into each other (Fig. 8A). After 4 weeks in culture, a nearly confluent monolayer of cells formed from the colonies. In contrast, after hMSCs were plated at low density, the colonies stopped expanding after 2-3 weeks, and even after 4 weeks the cultures did not grow to confluency (Fig. 8B).
The results here demonstrated that rMSCs are similar to hMSCs in that cultures contain at least three subpopulations of cells based on size and granularity. Also, low-density plating enhances expansion of both rMSCs and hMSCs. However, the results demonstrated that rMSCs differ from hMSCs in several ways: A) rMSCs are more sensitive to plating density; B) they expand more rapidly after low-density plating; C) they require frequent medium changing; D) they form confluent cultures after low-density plating, and E) the mature MSCs in cultures can generate RS cells and single-cell-derived colonies.
At the moment, it is difficult to explain the differences observed between rMSCs and hMSCs. One possibility is that the rMSCs underwent transformation as is frequently seen with mouse fibroblasts in cultures. However, preliminary karyotyping analyses of the cells have not revealed any marked chromosome aberrations. Another possible explanation is that initial samples of rMSCs contain more primitive progenitor cells than samples of hMSCs since the rMSCs are obtained by thorough flushing of long bones, whereas hMSCs are obtained by random sampling of marrow using a needle and syringe. However, the differences in harvesting procedures are unlikely to explain all the observations here. Although not readily explained, the differences between cultures of rMSCs and hMSCs will be important to consider in future experiments in which rMSCs are employed in the large number of rat models that are now available for human diseases.
The work was supported in part by NIH research grants AR47796 and AR44210, a gift from the Oberkotter Foundation (Philadelphia, PA), the Louisiana Gene Therapy Research Consortium (New Orleans, LA), and HCA—The Healthcare Company (Nashville, TN).