Full Text
Multilineage Potential of Adult Human Mesenchymal Stem Cells
Mark F. Pittenger, Alastair M. Mackay, Stephen C. Beck, Rama K. Jaiswal, Robin Douglas, Joseph D. Mosca, Mark A. Moorman, Donald W. Simonetti, Stewart Craig, and Daniel R. Marshak

Supplementary Material

Supplemental Figure 1. Human mesenchymal stem cell (hMSC) isolation and characterization. (A) Karyotype analysis. The isolated hMSCs from three donors (2 male and 1 female) were tested at passage 12 by a commercial testing lab (Nichols Institute, San Juan Capistano, CA). All donors were found to be normal. (B) Passaged hMSCs retain their telomerase activity. The telomeric repeat amplification protocol (TRAP-eze) (Oncor, Gaithersburg, MD) assay was performed on cultured human fibroblasts and mesenchymal stem cells (MSCs). Telomerase activity was revealed by the generation of radiolabeled TRAP products [ladder of bands starting at 50 base pairs (bp)] by cell lysates (+ lanes) and not by heat-inactivated lysates (- lanes). The dark band at the bottom of the non-denaturing polyacrylamide gel is an internal control (36 bp). Telomerase activity was detected in both early and late passage hMSCs.
Isolation of hMSCs. Informed consent was obtained from volunteer donors. hMSCs were isolated from marrow of the iliac crest of normal adults by a modification of the procedure of S. E. Haynesworth, J. Goshima, V. M. Goldberg, A. I. Caplan, Bone13, 81 (1992). About 20 to 30 ml of marrow aspirate was collected into a syringe containing 6000 units of heparin to prevent clotting. The marrow sample was washed with Dulbecco's phosphate-buffered saline (DPBS), cells were recovered after centrifugation at 900g, and the process was repeated once more. Up to 2 ( 108 nucleated cells at 4 (107 cells/ml were loaded onto 25 ml of Percoll of a density of 1.073 g/ml in a 50-ml conical tube. Cell separation was accomplished by centrifugation at 1100g for 30 min at 20°C. The nucleated cells were collected from the interface, diluted with two volumes of DPBS, and collected by centrifugation at 900g. The cells were resuspended, counted, and plated at 200,000 cells/ cm2. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (low glucose) containing 10% fetal bovine serum (FBS) from selected lots. The serum lots selected for hMSC outgrowth from marrow aspirates were chosen for their ability to promote the growth of an adherent, well-spread colony-forming cell that, when placed into ceramic cubes and implanted into athymic mice, would give rise to bone and cartilage when evaluated histologically [D. P. Lennon et al., In Vitro Cell. Dev. Biol.32, 602 (1996). The experiments described here utilized hMSCs that were grown in FBS lot AFE5185 from Hyclone (Logan, UT). Medium was replaced at 24 and 72 hours and every third or fourth day thereafter. hMSCs grew as symmetric colonies and were subcultured at 10 to 14 days by treatment with 0.05% trypsin and 0.53 mM EDTA for 5 min, rinsed from the substrate with serum-containing medium, collected by centrifugation at 800g for 5 min, and seeded into fresh flasks at 5000 to 6000 cells/cm2. With each treatment of trypsin-EDTA and replating, the passage number was increased and represented approximately three population doublings.

Supplemental Figure 1. Table 1.

Average volume of marrow aspirate 10 ml
Number of nucleated cells 280 x 106/ml
Average number of cells after density
~100 x 106 (30%)
Cells plated at 200,000/cm2 3 flasks (185 cm2 size)
Average number of cells at passage 0
Replate cells at 1 x 106/flask
2 million to 5 million/flask = 6 million to 15 million
Average number of cells at passage 1
Replate cells at 1 x 106/flask
3 million to 5 million/flask = 18 million to 75 million
Average number of cells at passage 2 3 million to 5 million/flask = 54 million to 375 million

Flow cytometry. The cells were harvested from the tissue culture flasks by treatment with 0.05% trypsin or 25 mM EDTA in phosphate-buffered saline (PBS). The cells, in solution at a concentration of 0.5 x 106 cells/ml, were stained for 20 min with an empirically determined amount of each antibody, generally 10 to 20 μl. Labeled cells were thoroughly washed with two volumes of PBS and fixed in flow buffer (1% paraformaldehyde, 0.1% sodium azide, and 0.5% bovine serum albumin in PBS). The labeled cells were analyzed on a FACScan or FACSVantage (Becton-Dickinson) by collecting 10,000 events with the Cell Quest software program (Becton-Dickinson).

Supplemental Figure 1. Table 2. Flow cytometry characterization of hMSCs

Integrins - Positive Integrins - Negative
α1 CD49a α4 CD49d
α2 CD49b αL CD11a
α3 CD49c Cβ2 CD18
CD49e Hematopoietic Markers -Negative
αv CD51 T4 CD4
β1 CD29 Mo2 CD14
β3 CD61 CD34
β4 CD104 Leukocyte Antigen
Cytokine Receptors - Positive Cytokine Receptors - Negative
IL-1R CD121a IL-2R CD25
IL-3Rα CD123
IL-4R CDw124
IL-6R CD126
IL-7R CDw127
Factor Receptors - Positive Factor Receptors - Negative
TGFβIIR Other - Negative
bFGFR Fas ligand
Transferrin CD71
Matrix Receptors Positive Matrix Receptors - Negative
ICAM-2 CD102 E-Selectin CD62E
VCAM-1 CD106 P-Selectin CD62P
L-Selectin CD62L PECAM-1 CD31
LFA-3 CD58 vW Factor
ALCAM CD166 Cadherin 5
Hyaluronate CD44 Lewisx CD15
Endoglin CD105
Others - Positive
Thy-1 CD90

Supplemental Figure 2. Adipogenic differentiation of marrow-derived stem cells. (A) hMSCs were cultured as monolayers in DMEM containing 10% FBS and antibiotics and allowed to become confluent. The cells were cultured for an additional 3 to 7 days and then adipogenic induction (MDI+I) medium containing 1 μM dexamethasone and 0.5 mM methyl-isobutylxanthine, insulin (10 (g/ml), 100 mM indomethacin, and 10% FBS in DMEM was added. The hMSCs were incubated in this media for 48 to 72 hours, and the media was changed to adipogenic maintenance (AM) medium containing insulin (10 μg/ml) and 10% FBS in DMEM for 24 hours. Lipid vacuoles were first detectable within the cells at 48 hours of the first MDI+I treatment. The cells were then re-treated with MDI+I for a second or third treatment. The cultures were then maintained in AM for 1 week before fixation. Adipogenic differentiation was demonstrated by the accumulation of lipid vesicles and by the expression of adipose-specific genes. Multiple treatments resulted in increasing numbers of adipocytes, as shown by oil red O staining (A). (B) A fluorescence assay based on Nile red staining in the lipid vacuoles and normalization relative to 4',6'-diamidino-2-phenylindole staining of DNA demonstrated the 9- to 16-fold increase over control cultures at 2 weeks. (C) Adipose-associated gene products were shown by immunoblotting, using an antibody to aP2 or using a cDNA probe for lipoprotein lipase (LPL) or peroxisome proliferator-activated receptor γ2 (PPARγ2) [Murine cDNA probes for LPL and PPARγ2 were the gift of J. Gimble (University of Oklahoma, School of Medicine, Oklahoma City), and polyclonal antibody to aP2 was provided by M. D. Lane (Johns Hopkins University, School of Medicine, Baltimore, MD).] Immunoblotting analysis of extracts of the adipogenic cells showed high levels of expression of aP2 [(C), adipogenesis (Adipo) lane], but undifferentiated or nonadipogenic cells that were present in the same culture [such as those present in 2 x MDI+I in (A)] failed to express detectable aP2 (control and non-Adipo lanes). LPL was present as two mRNAs of 3300 and 3700 nucleotides (nt), and PPARγ was present as a single mRNA of ~1800 nt. Northern (RNA) blots of extracts of the adipogenic cells showed high-level expression of PPARγ and LPL but showed no expression in the undifferentiated cultured mesenchymal cells [(C), lower panels]. (D) Lipid vesicles in adipocytes coalesced over 3 months in culture to form one or two large lipid inclusions.

Supplemental Figure 3. Chondrogenic differentiation of marrow-derived stem cells. Cultured marrow-derived stem cells were grown as a pelleted micromass with the inclusion of TGFβ3. Under low-speed centrifugation, a dense mass of cells formed at the bottom of the conical centrifuge tube. The cells consolidated within 1 day to a single mass that could be dislodged to float freely in suspension culture. Safronin O staining for proteoglycans revealed an increasing amount over the 3-week time period associated with the morphological differentiation to a chondrocyte phenotype: (A) 7 days, (B) 14 days, (C) 21 days, and (D) 28 days. The differentiated cells were also highly reactive with C4F6 monoclonal antibody to type II collagen. The chondrogenic cells could be induced to further differentiate to hypertrophic chondrocytes that expressed (E) type X collagen (monoclonal X53 from Quartett, Berlin) as well as (F) type II collagen (monoclonal C4F6). The chondrogenic cells also expressed (G) the extracellular matrix proteoglycan aggrecan (polyclonal antiaggrecan from D. Heinegard, Lund University, Lund, Sweden). The cell pellet increased in size over a 2- to 3-week period, and this is due almost entirely to the deposition of extracellular matrix rather than continued cell division [A. M. Mackay, S. C. Beck, J. M. Murphy, F. P. Barry, M. F. Pittenger, Tissue Eng.4, 472 (1998)].

Supplemental Figure 4. Osteogenic differentiation of marrow-derived stem cells. (A) Osteogenic differentiation of marrow-derived stromal cells over a period of 21 days was demonstrated by the increase in alkaline phosphatase and the accumulation of calcium, which has been shown to be in the form of crystalline hydroxyapatite [N. Jaiswal, S. E. Haynesworth, A. I. Caplan, S. P. Bruder, J. Cell. Biochem.64, 295 (1997)]. Alkaline phosphatase was detected histologically (Sigma kit 85), and the mineral was stained with silver by the method of von Kossa. (B) The quantification of alkaline phosphatase demonstrated an increase in activity between 7 and 14 days that then subsided, whereas (C) that for calcium indicated that the accumulation continued to at least 3 weeks.

Supplemental Figure 5. Differentiation is restricted to the directed lineage as shown by reverse transcriptase-polymerase chain reaction (RT-PCR). RT-PCR was used to analyze the expression of lineage-related mRNAs and to reveal whether any cells differentiated to the unintended lineages. hMSCs were cultured in each differentiation condition for up to 2 weeks, and RNA was then isolated. At this time, most of the cells had differentiated to the desired lineage, and the RT-PCR reaction was performed and the resultant cDNA population was tested for expression of PPARγ, aP2, type II collagen, type IX collagen, osteopontin, and alkaline phosphatase (AP). Primers for β2 microglobulin (B2M) were used as a control to assure even loading of the gel. Cells were rinsed in DPBS, lysed in RLT lysis buffer (Qiagen, Santa Clarita, CA), total RNA was purified with Qiagen RNeasy spin columns according to the manufacturer's recommendations, and the isolated RNA was quantified by ultraviolet spectroscopy. For PCR analysis, 1 μg of RNA was converted to cDNA with Moloney murine leukemia virus RT and random hexamer primers. The reagents were purchased from Perkin-Elmer (Foster City, CA), and experiments were performed with a Perkin-Elmer GeneAmp 9600 PCR system and MicroAmp reaction tubes. The PCR reactions were carried out for 30 cycles with the primers given below. The reaction products were resolved by electrophoresis on a 1.5% agarose gel and visualized with ethidium bromide. PCR primers were as follows:
PPARγ2 (352 bp product)
Fatty acid binding protein aP2 (114-bp product)
Type II collagen 1α (377-bp product)
Type IX collagen α1 (159-bp product)
Osteopontin (330-bp product)
β2 Microglobulin (270-bp product)
PPARγ2 and aP2 were only expressed at detectable levels in the adipogenic cultures, whereas type II and type IX collagens were only found in the chondrogenic conditions. Osteopontin was expressed in the osteogenic and chondrogenic pathways, with low-level expression in the undifferentiated hMSCs (Control). Alkaline phosphatase was expressed in all preparations, but was highest in the osteogenic hMSCs. Peroxisome proliferator-activated receptor (PPAR) γ2 and the fatty acid binding protein aP2 were only expressed in the adipogenic cultures, whereas type II and type IX collagens were only expressed in the chondrogenic cultures. Osteopontin was found in chondrogenic as well as in osteogenic cultures, which is consistent with in vitro differentiation of these tissues. Alkaline phosphatase was present in all cultures but was most elevated in the osteogenic cultures. Cbfa-1 expression, as tested by RT-PCR, was not a good marker for osteogenesis, as it was expressed in the undifferentiated cells and expression persisted in cells differentiated to the other lineages (R. K. Jaiswal and M. F. Pittenger, data not shown). β-2-microglobulin was used as a control for equal loading in these experiments. Additional information on hMSCs is available at http://www.osiristx.com.