Trans-differentiation of stem cells shows promise for use in tissue repair medicine. Although poorly defined, mesenchymal stem cells (MSC) appear useful for applications in repair medicine. Despite the low frequency of MSC, they are relatively easy to expand. The expression of MHC class II on MSC, however, could deter their use in repair medicine, since these molecules could stimulate an allogeneic host response. This study sought to compare the immune stimulatory and suppressive effects of MSC. Primary human MSC were cultured from bone marrow aspirates and then passaged at least three times before use in assays. Morphologically, MSC were symmetrical; were SH2+, MHC class II+, CD45, CD44+, CD31, CD14, proly-4-hydroxylase; and showed normal karyotype patterns and elevated telomerase activities. MSC elicited significant stimulatory responses when cocultured with allogeneic PBMC. Despite the production of different types of growth factors, allogeneic effects of MSC could not be explained by the production of these growth factors. One-way MLR reactions were significantly blunted by third-party MSC. Similar suppression was not observed for responses to three different recall Ags. Based on these functional differences by MSC in responses to allo- and recall Ags, we examined whether MSC could exert veto-like functions. We showed that MSC could blunt the cytotoxic effects of allogeneic-induced effectors to mitogen-activated targets. The results showed that although MSC elicited allogeneic responses in a model that mimics a graft-vs-host reaction, they also exerted veto-like activity, but caused no effect on responses to recall Ags.

The literature suggests that the adult bone marrow (BM)3 might host several types of stem cells (1, 2, 3, 4). Lymphohemopoietic (LHSC) and mesenchymal (MSC) stem cells have been shown to reside in the BM (1, 5), with frequencies ranging between 0.01 and 0.001% of BM cells (6). LHSC and MSC are morphologically, phenotypically, and functionally distinct (7, 8). LHSC are mostly found distant from the marrow sinus in the areas with the lowest oxygen levels (9, 10, 11). MSC surround the vasculature and project toward the cellular areas of the BM microenvironment (2, 5). LHSC are sources of immune cells and are therefore considered to be the cells of the emerging immune system in the adult (1, 8). The stromal cell support of LHSC could be derived from differentiated MSC (5). Thus, it could be argued that MSC support the self-renewal and repopulation functions of LHSC.

LHSC can trans-differentiate into nonhemopoietic cells (12, 13). MSC similarly possess trans-differentiation potential and may prove to be clinically useful (3, 7). MSC might be the progeny of a more primitive mesenchymal stem cell, described as multiple adult progenitor cell (3, 14). Regardless, MSC and/or their precursor(s) might withstand the test of science in the long term as the best candidate stem cells for application in repair medicine (3, 15). MSC can differentiate into different cell lineages, including BM stroma (5). The ability of MSC to form stroma gives these cells the potential to be adjuvant cells in BM transplantation to improve engraftment of the LHSC (2, 16).

MSC show plasticity, based on their ability to trans-differentiate into cells of other germ layers (17). MSC have been reported to form neurons (18, 19, 20, 21), hepatocytes (14), muscle cells (22), and endothelia (23, 24). The lineage commitment and trans-differentiation properties of MSC make these stem cells candidates for use in repair/molecular medicine and tissue engineering. MSC have the potential to repair bone and could conceivably be used to replace osteoblasts in patients with osteogenesis imperfecta (2).

Morphologically, MSC could be confused with BM fibroblasts/stroma or at times even neurons. However, MSC are symmetrical, while stromal cells are asymmetrical. LHSC, and their progenies can be distinguished from MSC by expression of nonoverlapping markers. MSC are negative for CD34, CD45, and CD14 (16) and express endoglin (SH2 and SH3), CD29, CD44, CD71, CD90, CD106, and CD120a (7). Both endothelial cells and MSC express endoglin (25); however, despite the close anatomical association of MSC with the vasculature, MSC are negative for the endothelial cell marker, CD31 (7).

At this time it is uncertain how the potential expression of MHC class II might affect implantation or transplantation of MSC. Most of the cell types derived from trans-differentiated MSC will not express MHC class II, unless there is an inflammatory response. For example, neural cells can express MHC class II under certain pathophysiological conditions (26). Current studies do not provide definitive answers regarding the differentiation stage at which MSC will be transplanted into patients. The question that arises is whether MSC should be implanted in an injured tissue as partly or fully trans-differentiated cells or if chemotactic mechanisms should be applied to the site of injury to attract autologous MSC.

Low expression of MHC class II on MSC gives these cells the potential to elicit weak allogeneic responses. Thus, it is possible that MSC could mediate alteration of the immune system during the interval between transplantation and trans-differentiation. Reports have shown that MSC are low allogeneic stimulators and might even suppress an ongoing immune response (27). Therefore, the design of clinical strategies to use MSC in BM transplantation and repair medicine cannot ignore their allogeneic properties, which could be solved if MSC show properties of veto functions.

Veto cells suppress CTL precursor function against Ags present on the veto cell surface, but not against those on third-party allogeneic cells (28, 29). Veto cells have been shown to prevent the rejection of allogeneic BM grafts in an Ag-specific manner (30). Veto cells are ubiquitously distributed, including in the BM (28, 29, 30, 31). In the present study we showed immune-suppressive effects of MSC during allogeneic responses and also demonstrated veto-like properties of MSC. This report explores questions that provide further insight into the biology of MSC, and the results have important implications about their clinical utility.

BM aspirates and peripheral blood (10–20 ml) were taken from healthy subjects ranging in age from 20 to 30 years. BM aspirates were taken from the posterior iliac crest and immediately placed in medium containing preservative-free heparin. The protocol for human subjects adhered to guidelines outlined by the institutional review board of the University of Medicine and Dentistry of New Jersey (Newark, NJ). Allogeneic differences of peripheral blood and BM donors were determined in one-way MLR, described below.

The following were purchased from Sigma-Aldrich (St. Louis, MO): sodium heparin, dexamethasone, β-glycerophosphate, silver nitrate, Ficoll-Hypaque, paraformaldehyde (PFA), DMEM with high glucose, RPMI 1640, BSA, α-MEM, DMEM, Con A, LPS, hydrocortisone, and glutamine. FCS and horse serum (HyClone, Logan, UT) were heat-inactivated for 45 min at 56°C. PBS, PHA, and colcemid were purchased from Life Technologies (Grand Island, NY). l-ascorbic acid-2-phosphate was purchased from Wako Chemicals (Osaka, Japan).

The following Abs were purchased from BD PharMingen (San Jose, CA): PE-conjugated HLA-DR mAb, PE-conjugated CD14 mAb, PE-rat anti-mouse κ, polyclonal FITC-conjugated anti-mouse IgG, isotype control IgG, and FITC-conjugated CD45 mAb. CD44 mAb was purchased from T Cell Diagnostics (Cambridge, MA). CD31 and prolyl-4-hydroxylase mAbs were purchased from DAKO (Carpinteria, CA). FITC-conjugated nonimmune rabbit IgG was purchased from Sigma-Aldrich.

The SH2 hybridoma (HB-10743) was purchased from American Type Culture Collection (Manassas, VA) and grown according to their instructions. SH2 cells were injected i.p. in BALB/c mice that were >6 mo old. Mice were housed in an American Association for Accreditation of Laboratory Animal Care-accredited animal facility, and the use of mice for propagation of ascites was approved by the University of Medicine and Dentistry of New Jersey institutional animal care and use committee. SH2 Abs were collected as ascites from mice and then purified by affinity on a Sepharose G column. Purified IgG was tested against MSC in dose-binding immunofluorescence assays. Nonspecific binding was determined with BM fibroblasts. For immunofluorescence assays, SH2 mAb was used at 50 ng/ml of purified IgG.

Immediately after taking BM aspirates from a donor, 2 ml was added to 15 ml of DMEM with 10% FCS in Falcon 3003 petri dishes (Fisher Scientific, Springfield, NJ). Plates were incubated at 37°C for 3 days, after which RBC and granulocytes were removed by Ficoll-Hypaque density-gradient centrifugation. Cultures were reincubated until confluence, which occurred at ∼2–3 wk after removal of RBC. MSC are sensitive to the toxic effects of Ficoll-Hypaque and were quickly collected and replaced in the cell culture flask following density-gradient centrifugation. FCS is crucial for optimal propagation of MSC and required proper quality control screening before its use to culture MSC. The FCS used in our studies ensured stable phenotype expression by cultured MSC for more than seven passages. Cells were passaged at least three times before being used in the assays (described below). Cells used in all assays were negative for prolyl-4-hydroxylase, CD14, CD31, and CD45 and positive for SH2 and CD44 (32) (Table I).

Table I.

Immunofluorescence for cell-specific protein in MSC and BM fibroblastsa

AgMSCFibroblasts
SH2 − 
CD45 − − 
CD44 − 
CD31 − ND 
CD14 − ND 
Fibroblast   
 Prolyl-4-hydroxylase − 
Neuronal   
 NeuN − − 
 Nestin − − 
 MAP2 − − 
AgMSCFibroblasts
SH2 − 
CD45 − − 
CD44 − 
CD31 − ND 
CD14 − ND 
Fibroblast   
 Prolyl-4-hydroxylase − 
Neuronal   
 NeuN − − 
 Nestin − − 
 MAP2 − − 
a

Confluent BM MSC and fibroblasts were labeled with specific antibodies and then scored (+ or −) using fluorescence microscopy or flow cytometry. Data were compiled from four experiments, each of which was performed with a different BM donor.

Stromal cultures were established with unfractionated BM aspirates as previously described (33). Briefly, BM aspirates were placed in α-MEM containing 12.5% FCS, 12.5% horse serum, 10−7 M hydrocortisone, 10−4 M 2-ME, and 1.6 mM glutamine. After 3 days at 33°C, RBC and granulocytes were removed by Ficoll-Hypaque density gradient centrifugation, and cultures were reincubated with weekly replacement of 50% medium until cells reached confluence. BM fibroblasts were prepared by passaging trypsinized BM stroma at least five times in α-MEM with 20% FCS. Fibroblast cultures were examined for purity by immunofluorescence and cytochemical staining as previously described (33).

PBMC were isolated by Ficoll-Hypaque density gradient centrifugation of peripheral blood. PBMC were washed and then resuspended at 2 × 106/ml in RPMI 1640 with 10% FCS and 5 × 10−7 M 2-ME (R10 medium). MSC were deadhered by trypsinization or gentle scraping with a disposable cell scraper. PBMC and MSC were resuspended in R10 medium at 2 × 106 and 105/ml, respectively. Stimulators (PBMC and MSC) were gamma-irradiated (200 cGy) with a cesium source. After irradiation cells were washed once, recounted, and then resuspended at the same concentrations, as stated above. Dose-response assays determined 105 MSC/ml to be the optimal concentration for one-way MLR with MSC stimulators.

MLR were performed in 96-well, flat-bottom plates (Corning, Corning, NY). Equal volumes (0.1 ml) of stimulators (gamma-irradiated PBMC or MSC) and responders (PBMC) were added to each well, and cultures were performed in triplicate. The cultures were pulsed with 1 μCi/well of [methyl-3H]TdR (70–90 Ci/mmol; NEN, Boston, MA) during the last 16 h of a 4-day culture. Cells were harvested with a PhD cell harvester (Cambridge Technologies, Cambridge, MA) onto glass-fiber filters (Cambridge Technologies). [3H]TdR incorporation was quantified in a scintillation counter (Beckman, Fullerton, CA). The results are expressed as the stimulation index (S.I.), which is the mean dpm of experimental cultures (responders + gamma-irradiated stimulators)/dpm of responder cells with only medium.

Two modifications of MLR were performed. 1) MSC were fixed with 0.4% PFA, and 104 cells were used as stimulators with PBMC responders. PFA fixation was performed by resuspending 104 MSC/ml in 0.4% PFA (solution prepared in PBS, pH 7.4) for 5 min. After this, cells were washed twice in PBS and then resuspended in R10 medium. 2) One-way MLR was performed with nonirradiated PBMC as responders and gamma-irradiated allogeneic PBMC as stimulators. Parallel cultures contained third-party allogeneic (nonirradiated or gamma-irradiated) MSC or BM fibroblasts. MSC and BM fibroblasts were cultured from BM aspirates of the same donor. Allogeneic differences among BM donors as well as donors of responder and stimulator PBMC were pretested in the one-way MLR (described above).

51Cr release assay.

For use as target cells, PBMC from healthy donors were resuspended at 107 cells/ml in R10 medium, and 1 ml was added to 4 ml of R10 medium in 25-cm2 tissue culture flasks (Costar; Corning). Cells were stimulated with PHA (final dilution, 1/1000), and flasks were incubated in an upright position. On day 5 cells were washed three times, and 107 cells in 1 ml of R10 medium were labeled with 200 μCi of 51Cr (401 mCi/mg; NEN) for 2 h. Cells were washed five times and then resuspended in R10 medium at 105 cells/ml for use in the cytotoxicity assay.

Cytotoxic effector cells were prepared to ascertain that the cells were primed against the alloantigens of the donors’ target cells (described above). Therefore, effector PBMC showed allogeneic differences from target PBMC. Stimulators (106/ml) were irradiated with 200 cGy and cocultured with an equal number of nonirradiated responder PBMC in upright 25-cm2 tissue culture flasks (5 × 106 cells total/flask) at 37°C for 5 days.

Cytotoxic assay.

Effector cells were harvested and then washed three times. Viable cells were resuspended at 107 cells/ml in R10 medium. Assays were performed in 96-well, round-bottom, tissue culture plates (Costar) at E:T cell ratios of 100:1, 50:1, 25:1, and 12.5:1 with or without MSC (103–105 cells total). The following controls were performed in sextuplet: target cells incubated in medium alone to determine spontaneous release of 51Cr from targets, and target cells incubated in medium containing 1% Triton X-100 to determine maximal isotope release from targets. Typically, spontaneous release was <10%, and maximum release was >90% of the absolute counts per minute of target cells. After 4 h cell-free supernatants were collected, and the amounts of 51Cr released through cell lysis were determined in a gamma counter. The percent cell lysis was calculated according to the following equation: (experimental point (dpm) − spontaneous release (dpm))/(total release (dpm) − spontaneous release (dpm)) × 100.

The human cytokine protein array II was purchased from Ray Biotech (Norcross, GA) and used according to the manufacturer’s instructions. Briefly, membranes were incubated for 30 min in 5% BSA in 0.01 M Tris buffer with 0.15 M NaCl, pH 7.6. After this, membranes were subjected to three 5-min washes with 1× TBS/0.1% Tween, followed by two 5-min washes with TBS. Fresh culture medium or cell-free supernatants from confluent MSC were added to the membrane, which were incubated for 1 h with biotin-conjugated cytokines (provided with the kit). After incubation, membranes were washed twice with TBS and then incubated for 30 min with HRP-conjugated streptavidin. Unbound reagents were washed, and the membranes were developed with the ECL system (NEN).

Immunofluorescence studies were performed microscopically and by FACScan. For microscopy, MSC or fibroblasts were cultured on coverslips that were placed in a 35-mm petri dish. At 80% confluence, nonadherent cells were aspirated, and the adherent cells washed with PBS, pH 7.4. Primary Abs were diluted in PBS with Tween and then added to coverslips. Cells were incubated with Abs for 30 min at room temperature, washed three times with PBS/Tween, and secondary Abs were added for 30 min. Control slides were incubated with secondary Ab alone, fluorescein-conjugated primary Ab alone, or a fluorescein-conjugated isotype control Ab. Cells were examined for fluorescence intensity using Probis microscope (Olympus, New Hyde Park, NY). For FACScan, cells were trypsinized and similarly labeled while in suspension.

MSC were seeded into Falcon 3003 petri dishes. After 12 h, cells were ∼40–50% confluent. At this time, 50 μl of colcemid (10 μg/ml) was added to each dish to yield a final concentration of 0.1 μg/ml, and cultures were reincubated overnight at 37°C. Cells were collected by trypsinization, washed, lysed with 0.75 M hypotonic KCl, and fixed with acid/alcohol (3/1). Chromosomes in metaphase were analyzed following Wright’s staining.

Telomerase activities from MSC and BM fibroblasts were determined with the Telo TAGGG telomerase PCR ELISA Plus kit (Roche, Indianapolis, IN). The technique followed the manufacturer’s instructions. Briefly, cells at ∼50–75% confluence were trypsinized and washed with PBS, pH 7.4, and 2 × 105 cells were lysed with the lysis reagent provided in kit. Cell-free supernatants were transferred to another tube and then kept on ice until 2 vol of lysate (3 and 10 μl) were analyzed in PCR for telomerase activity, which is proportional to the density of telomeric repeats.

Adipogenic differentiation was performed with the Adipogenic hMSC Differentiation Bullet Kit (Cambrex BioScience, Walkersville, MD). The method followed the manufacturer’s recommendations. Briefly, MSC at ∼80% confluence were trypsinized and then replated at 5 × 105 cells/well in six-well tissue culture plates. At cell confluence, adipogenic differentiation was performed with three cycles of induction/maintenance medium. Each cycle consisted of 3-day culture in adipogenic induction medium, followed by 1–3 days with adipogenic maintenance medium. Control wells were cultured, and medium was replaced on the same schedule using adipogenic maintenance medium. After the three-cycle schedule, cells were grown in adipogenic maintenance medium for 7 days, with change of medium every 3 days. After this, cells were washed with PBS, fixed with 10% formalin, stained with Oil Red O, and counterstained with hematoxylin.

MSC were trypsinized, replated at 2 × 104 cells/well of a six-well plate, and allowed to adhere for 24 h. Osteogeneic differentiation was induced in DMEM containing 10% FCS, 100 nM dexamethasone, 10 mM β-glycerophosphate, and 50 μM l-ascorbic acid-2-phosphate. Differentiation medium was replaced every 3–4 days for 3 wk. Controls wells were grown and replaced with MSC medium on the same schedule. Osteogenic differentiation was assessed by mineral deposits by von Kossa stain. Cells were fixed with 10% formalin, stained for 10 min in the dark with 2% silver nitrate solution, washed, and exposed for 15 min to light.

Data were analyzed using ANOVA and Tukey-Kramer multiple comparisons test. A value of p <0.05 was considered significant.

Fig. 1,A shows photomicrographs of MSC cultures taken at different growth confluencies. Morphologically, MSC are spindle-shaped and symmetrical with respect to the nuclei. The morphology of MSC was the same regardless of the degree of confluence or the number of cell passages (sparse growth, Fig. 1,A, left; confluent culture, Fig. 1,A, right). In our hands we could not alter the formulation of the culture medium used to grow MSC (described in Materials and Methods). A change in the culture medium resulted in cells that were morphologically different from those shown in Fig. 1,A. The symmetrical, linear morphology of MSC (Fig. 1,A) could be distinguished from the asymmetry and polarization in multiple directions exhibited by BM fibroblasts (Fig. 1 B). To differentiate BM fibroblasts (BM stroma) from MSC is relevant because BM fibroblasts are differentiated progenies of MSC (34).

FIGURE 1.

Characterization of MSC. A, Representative culture of MSC at different confluence. Magnification: left, ×40; right, ×20. B, MSC (left) and fibroblasts (right) from the same BM donor (magnification, ×20). C, Karyotype of human MSC after five passages. D and E, Representative of three experiments, each performed with a different donor. Stains for adipogenic (D) and osteogenic (E) differentiation are shown.

FIGURE 1.

Characterization of MSC. A, Representative culture of MSC at different confluence. Magnification: left, ×40; right, ×20. B, MSC (left) and fibroblasts (right) from the same BM donor (magnification, ×20). C, Karyotype of human MSC after five passages. D and E, Representative of three experiments, each performed with a different donor. Stains for adipogenic (D) and osteogenic (E) differentiation are shown.

Close modal

MSC (passages 1–5) were analyzed by flow cytometry with a panel of Abs specific for cells that can be confused (morphologically and by organ location) with MSC (Table I). The results from FACScan are represented as positive and negative (Table I). Since MSC are proposed as cells able to trans-differentiate into neurons (18, 19), the morphology of MSC (Fig. 1,A, left panel) could be confused with that of neurons. We therefore labeled MSC from four different donors with Neu-N, microtuble-associated protein-2, and nestin and showed negative staining (Table I). The results shown in Table I for passage 3 are consistent with the literature on human MSC (7).

We previously established cultures of BM fibroblast with mononuclear cells from BM aspirates and α-MEM containing 20% FCS (33). With recent phenotypic definition of MSC (7), we re-evaluated our technique for culturing fibroblasts (7) and found <5% contamination of cells that are positive for SH2 (not shown). We now show that purified BM fibroblasts (no detection of SH2+ cells; Table I) can be obtained by subculture with five passages of confluent BM stromal cells (35). Immunofluorescence staining of these fibroblasts at passage 5 showed that in addition to a distinct morphology (Fig. 1,B), fibroblasts could be distinguished from MSC by immunofluorescence analyses (Table I).

Multiple passages of MSC and in vitro manipulation of BM cells could cause changes in chromosomes, such as whole or partial and/or chromosomal fusion. Therefore, we performed cytogenetic analyses on representative MSC cultures. For each experiment the karyotype was normal (representative karyotype shown in Fig. 1 C). High telomerase activity is characteristic of stem cells (36). We therefore studied telomerase activities in MSC and in differentiated cells of mesenchymal origin, i.e., fibroblasts. In seven analyses, using a different donor, the results show 4-fold increases in telomerase activity in MSC compared with passaged BM fibroblasts. The high telomerase activity demonstrated in MSC is consistent with their stem cell property (36).

Osteogenic and adipogenic differentiation were performed as further verification that the cells shown in Fig. 1,A have stem cell properties (7). In three differentiation studies (each with a different donor), the cultures stained positive for Oil Red O (Fig. 1,D) and von Kossa stain (Fig. 1 E), indicating that the cells have adipogenic and osteogenic potential.

The level of MHC class II expression on MSC is undefined (2, 7). Since MSC are proposed as a potential source for tissue regeneration (3, 24), clarification of MHC class II expression is important. Therefore, we studied the expression of MHC class II by FACS analysis with purified MSC labeled with PE-conjugated HLA-DR mAb. The results showed a shift in >70% MSC to the right after adjustment with the isotype control (overlapping region of Fig. 2,A). Based on the broad peak shown in the fluorescence labeling (Fig. 2,A), we next studied the expression of HLA-DR at a single-cell level by immunofluorescence microscopy. The results (Fig. 2 B) showed bright fluorescence at the cell membrane. The fluorescence was observed in a punctated distribution pattern in areas close to the nuclei, but dim to undetectable on the projections that were distant from the nuclei. Isotype control staining did not show any fluorescence (not shown). The data show that MSC express HLA-DR, and that the expression is concentrated in areas of the cell membrane close to the nuclei.

FIGURE 2.

MHC class II expression on MSC. A, MSC were cultured to 20% confluence and then labeled with PE-conjugated HLA-DR mAb (magnification, ×40). B, FACS analyses of MSC labeled with PE-HLA-DR mAb or PE-isotype control.

FIGURE 2.

MHC class II expression on MSC. A, MSC were cultured to 20% confluence and then labeled with PE-conjugated HLA-DR mAb (magnification, ×40). B, FACS analyses of MSC labeled with PE-HLA-DR mAb or PE-isotype control.

Close modal

MSC have been reported to exhibit a capacity to stimulate only meager responses in allogeneic PBMC (27). However, since our results showed dense expression of MHC class II expression on MSC (Fig. 2), our next sets of studies were designed to determine whether MSC could elicit allogeneic responses. This question was addressed in one-way MLR with gamma-irradiated MSC as stimulators obtained from five different BM donors. Stimulator MSC from each donor was cultured with responder PBMC from seven different donors. One-way MLR verified that stimulators and responders were allogeneic with each other (not shown). Each bar in Fig. 3 represents the mean S.I. ± SD of seven responders (each from a different donor) with gamma-irradiated MSC from one donor. The results showed that MSC could elicit significant proliferation of responder PBMC with S.I. values ranging between 3.5 and 6.5.

FIGURE 3.

One-way MLR with gamma-irradiated MSC from five different donors. MSC from each donor were used as stimulator cells with PBMC responders from seven different donors. Details of the techniques are described in Materials and Methods. The results are expressed as the S.I. ± SD. Background disintegrations per minute (PBMC with medium alone) ranged from 136–188.

FIGURE 3.

One-way MLR with gamma-irradiated MSC from five different donors. MSC from each donor were used as stimulator cells with PBMC responders from seven different donors. Details of the techniques are described in Materials and Methods. The results are expressed as the S.I. ± SD. Background disintegrations per minute (PBMC with medium alone) ranged from 136–188.

Close modal

MSC have been reported to express genes for different growth factors (21, 34). Gamma-irradiated stimulator MSC could produce soluble factors in one-way MLR (Fig. 3) because the cells are metabolically activity. Thus, to determine whether cytokines produced by MSC were involved in the allogeneic responses of MSC (Fig. 3), we first studied the production of cytokines from MSC by microarray analysis. The microarray membranes, which comprised 43 different growth factors, were hybridized with supernatants from confluent MSC or fresh culture medium. Detectable spots were scanned by densitometer, and detection with fresh culture media was subtracted from the test samples. Positive controls, provided by the manufacturer, were normalized to 3-fold, and the unknown samples were calculated from the normalized spots. The results (Table II) are therefore presented as fold increase from the positive controls. High densities of growth factors belonging to the chemokine family and those associated with angiogenesis were detected. There was no obvious difference between pro- and anti-inflammatory cytokines. The results showed constitutive expression of growth factors that could be grouped into different categories based on hemopoietic functions, angiogenic properties, and immune regulation.

Table II.

Cytokine microarray with supernatants from confluent MSCa

Growth FactorsFold Increase (positive control/unknown)
Chemokines  
 RANTES, GRO-α 
 MCP-1, IL-8 
 MIP-1β, SDF-1α ND 
Cytokines  
 Broad-acting  
  IL-6 
 Proinflammatory  
  IL-2, IL-12 ND 
  IFN-γ, TNF-α, IL-1α 
 Anti-inflammatory  
  IL-4, IL-10, IL-13 ND 
  TGF-β 
 Hemopoietic  
  G-CSF, GM-CSF 
Angiogenic  
 Angiogenin 
 Oncostatin M 
Growth FactorsFold Increase (positive control/unknown)
Chemokines  
 RANTES, GRO-α 
 MCP-1, IL-8 
 MIP-1β, SDF-1α ND 
Cytokines  
 Broad-acting  
  IL-6 
 Proinflammatory  
  IL-2, IL-12 ND 
  IFN-γ, TNF-α, IL-1α 
 Anti-inflammatory  
  IL-4, IL-10, IL-13 ND 
  TGF-β 
 Hemopoietic  
  G-CSF, GM-CSF 
Angiogenic  
 Angiogenin 
 Oncostatin M 
a

Cell-free supernatants were taken from confluent MSC of three different BM donors and then analyzed by microarray. Results are from three of the samples. The densitometric scans of the internal positive controls were normalized to 3-fold, and the unknowns calculated accordingly. Details of the procedure are described in Materials and Methods. MCP-1, monocyte chemoattractant protein 1; MIP-1β, macrophage inflammatory protein; SDF-1α, stromal-derived factor.

The next set of studies determined whether the allogeneic responses shown in Fig. 3 could be due to soluble factors produced by MSC (Table II). This question was addressed by comparing MLR using paraformaldehyde-treated vs gamma-irradiated MSC as stimulators. The proliferative responses of MLR performed with paraformaldehyde-treated and gamma-irradiated stimulators were not significantly (p > 0.05) different (Fig. 4 A). Furthermore, the addition of supernatants obtained from paraformaldehyde-treated cells cultured for up to 24 h showed no effect on cell proliferation in one-way MLR with PBMC from two different donors (not shown).

FIGURE 4.

Gamma-irradiated MSC vs PFA-fixed MSC as allostimulators and effects of soluble factors produced by confluent MSC in one-way MLR. A, MLR with PFA-fixed MSC or gamma-irradiated MSC using techniques as described in Fig. 2. B, Supernatants from confluent MSC or fresh medium were added to MLR at 10% (v/v). Results showed the effects of supernatant from five different one-way MLR; each assay was performed with two different responders (PBMC) and three different MSC stimulators. The responses performed with fresh medium were normalized to 1, and the effects of supernatants from MSC are shown as the fold increase in cell proliferation. The results are show as the mean ± SD (n = 30). Background disintegrations per minute (PBMC with medium alone) ranged from 136–188. ∗, p > 0.05 vs gamma-irradiated MSC.

FIGURE 4.

Gamma-irradiated MSC vs PFA-fixed MSC as allostimulators and effects of soluble factors produced by confluent MSC in one-way MLR. A, MLR with PFA-fixed MSC or gamma-irradiated MSC using techniques as described in Fig. 2. B, Supernatants from confluent MSC or fresh medium were added to MLR at 10% (v/v). Results showed the effects of supernatant from five different one-way MLR; each assay was performed with two different responders (PBMC) and three different MSC stimulators. The responses performed with fresh medium were normalized to 1, and the effects of supernatants from MSC are shown as the fold increase in cell proliferation. The results are show as the mean ± SD (n = 30). Background disintegrations per minute (PBMC with medium alone) ranged from 136–188. ∗, p > 0.05 vs gamma-irradiated MSC.

Close modal

Since different growth factor proteins were produced by confluent MSC (Table I), we explored the possible influence of these growth factors in one-way MLR in which both stimulators and responders were PBMC. In five different one-way MLR studies (Fig. 4,B), we observed significant increases (p < 0.05) in cell proliferation with optimal volumes (10%, v/v) of MSC supernatants. These results indicate that growth factors collected from confluent cultures of MSC enhance the proliferation of the responding cells in one-way MLR in which both responders and stimulators are PBMC (Fig. 4,B). However, the supernatants produced by MSC are not responsible for the proliferation of allogeneic PBMC when gamma-irradiated MSC are stimulators (Figs. 3 and 4 A).

An in vitro model of on-going immune responses was used to further explore the dichotomy in the functions of MSC, i.e., their ability to stimulate allogeneic MLR (Fig. 3) vs their reported immunosuppressive activities (27). MSC were used as third-party cells in one-way MLR containing PBMC as responders and allogeneic gamma-irradiated PBMC as stimulators. Fig. 5 A shows the responses of PBMC from four different donor/responder combinations (A–D). Each bar represents cultures of one responder and four allogeneic PBMC stimulators in the presence or the absence of MSC from three different donors (gamma-irradiated or nonirradiated). Analyses of the results indicate an ∼3-fold decrease in cell proliferation when MSC were added as third-party cells. The effects were the same when irradiated or nonirradiated MSC were used (not shown). Furthermore, when supernatants from confluent MSC (in lieu of cells) were added to allogeneic MLR, they did not suppress the reaction (not shown).

FIGURE 5.

Effects of MSC or BM fibroblasts as third-party cells in one-way MLR. Responder (R) PBMC were stimulated with allogeneic stimulator cells (S) in the presence or the absence of gamma-irradiated MSC (A) or gamma-irradiated BM fibroblasts (B). Cell proliferation was determined as described in Materials and Methods. Each experimental point represents one responder (R) with four allogeneic stimulators (S) in the presence or the absence of BM MSC or fibroblasts. The data are presented as the mean ± SD. Background disintegrations per minute (PBMC with medium alone) ranged from 213–256. ∗, p < 0.05 vs gamma-irradiated MSC; ∗∗, p > 0.05 vs gamma-irradiated BM fibroblasts.

FIGURE 5.

Effects of MSC or BM fibroblasts as third-party cells in one-way MLR. Responder (R) PBMC were stimulated with allogeneic stimulator cells (S) in the presence or the absence of gamma-irradiated MSC (A) or gamma-irradiated BM fibroblasts (B). Cell proliferation was determined as described in Materials and Methods. Each experimental point represents one responder (R) with four allogeneic stimulators (S) in the presence or the absence of BM MSC or fibroblasts. The data are presented as the mean ± SD. Background disintegrations per minute (PBMC with medium alone) ranged from 213–256. ∗, p < 0.05 vs gamma-irradiated MSC; ∗∗, p > 0.05 vs gamma-irradiated BM fibroblasts.

Close modal

To exclude the possibility that the suppressive effects were not due to the increased cell density of adding MSC as third-party cells to allogeneic MLR, we next determined whether gamma-irradiated fibroblasts, obtained from the same BM donor, elicited similar effects as MSC (Fig. 5,A). Experiments similar to those described for Fig. 5,A were performed in the presence of fibroblasts obtained from the three donors used to prepare MSC in Fig. 5,A. In four different one-way MLR, the results showed no significant difference (p > 0.05) in cell proliferation in the presence or the absence of BM fibroblasts (Fig. 5,B), although a trend toward higher incorporation of [3H]TdR was observed when BM fibroblasts were added to cultures (Fig. 5,B, ▩). Each bar in Fig. 5 B represents the mean S.I. ± SD of one PBMC responder and four gamma-irradiated PBMC stimulators. Allogeneic differences between responders and stimulators were verified in one-way MLR (data not shown). Together the results indicate that while third-party MSC suppress one-way allogeneic MLR, similar suppression was not observed when BM fibroblasts from the same donor were added to the MLR.

MSC as third-party cells in MLR mediated inhibitory effects on cell proliferation (Fig. 5 A). To further clarify the effects of MSC as an immune suppressor, we added MSC to an assay in which PBMC were sensitized to recognize and produce cytotoxic effects or cells against stimulator cell alloantigens. Presensitized PBMC (effector) were established in 5-day cocultures with gamma-irradiated PBMC from an unrelated donor. Effectors from these MLR were analyzed for cytotoxicity against blast-activated targets, which were obtained from 3-day cultures with PHA.

Cytotoxic assays were performed with 104 target cells and variable number of effector cells necessary to obtain different E:T cell ratios. Cultures were established in the presence or the absence of constant numbers (103 cells) of MSC, and 51Cr released from targets was used as a readout of cytotoxicity. The percent 51Cr release at an E:T cell ratio of 25:1 (Fig. 6,A) indicated significant suppression in wells containing MSC (p < 0.05). However, suppression was not observed in wells containing a similar number of BM fibroblasts that were cultured from the donor of MSC (Fig. 6,A). The data in Fig. 6,B show a typical cytotoxic response in the absence of MSC for one donor at different E:T cell ratios. As shown in Fig. 6 C, the donor cells were verified for allogeneic differences in parallel studies with one-way MLR. Based on the S.I. shown, it could be deduced that the donors used to induce cytotoxicity (effectors and targets) were significantly different with respect to allogeneic variations. It should be noted that allogeneic variation was considered if the donors of effector/target pairs showed an S.I. >3 in one-way MLR. These results indicated that MSC could also suppress the cytotoxic effects of Ag-primed effector cells in a short term (4-h) assay.

FIGURE 6.

Veto-like effects by MSC. Cytotoxic assay was performed with different ratios of 51Cr-labeled targets (T) and effectors (E). The target population comprised PHA-stimulated PBMC, and effectors used PBMC stimulated for 5 days with gamma-irradiated PBMC of target donors. Cytotoxic reactions were performed in the presence or the absence of MSC or BM fibroblasts. The latter cell types were derived from BM aspirates taken from unrelated donors. The results are expressed as the percent cell lysis at an E:T cell ratio of 25:1 (A). Representative cytotoxic reactions are shown for different E:T cell ratios (B). Allogeneic differences among donors of effectors, targets, and MSC are presented for eight experiments using one-way MLR with PBMC. Each experimental point represents the S.I. of one reaction (C). ∗, p < 0.05 vs cultures without gamma-irradiated MSC. The average disintegrations per minute for the total counts of targets was 30,335 ± 2,200.

FIGURE 6.

Veto-like effects by MSC. Cytotoxic assay was performed with different ratios of 51Cr-labeled targets (T) and effectors (E). The target population comprised PHA-stimulated PBMC, and effectors used PBMC stimulated for 5 days with gamma-irradiated PBMC of target donors. Cytotoxic reactions were performed in the presence or the absence of MSC or BM fibroblasts. The latter cell types were derived from BM aspirates taken from unrelated donors. The results are expressed as the percent cell lysis at an E:T cell ratio of 25:1 (A). Representative cytotoxic reactions are shown for different E:T cell ratios (B). Allogeneic differences among donors of effectors, targets, and MSC are presented for eight experiments using one-way MLR with PBMC. Each experimental point represents the S.I. of one reaction (C). ∗, p < 0.05 vs cultures without gamma-irradiated MSC. The average disintegrations per minute for the total counts of targets was 30,335 ± 2,200.

Close modal

The effects of MSC on Ag-specific cytotoxic responses (Fig. 6) suggested a veto-like function of MSC. Since veto cells should not be able to alter stimulatory effects of recall Ags, the next set of studies was performed to determine whether MSC could affect cell proliferation induced by common recall Ags. PBMC were stimulated with three different recall Ags (Candida albicans, Bordetella pertusis, and tetanus toxoid) in the presence or the absence of allogeneic or autologous MSC. Dose-response curves determined the optimal concentrations of each Ag. Suppressive effects by MSC on 7-day stimulation with the recall Ags was not observed (Fig. 7). The increased proliferation following the addition of MSC to allogeneic PBMC (Fig. 7) was expected, since MSC were previously shown to stimulate MLR when added to naive allogeneic PBMC (Fig. 3). In contrast, MSC showed no effect on proliferation when they were added to recall Ag-stimulated autologous PBMC (not shown). The results indicated that compared with the suppressive effects by MSC on allogeneic MLR and alloantigen-induced cytotoxic responses, they had no effect on the response to recall Ags.

FIGURE 7.

Effects of MSC on cell proliferation to recall Ags. PBMC were stimulated with optimal concentration of three different recall Ags in the presence of allogeneic MSC. The results are the mean ± SD of four different experiments. Background disintegrations per minute (PBMC with medium alone) ranged from 210–321. ∗, p < 0.05 vs cultures without gamma-irradiated MSC.

FIGURE 7.

Effects of MSC on cell proliferation to recall Ags. PBMC were stimulated with optimal concentration of three different recall Ags in the presence of allogeneic MSC. The results are the mean ± SD of four different experiments. Background disintegrations per minute (PBMC with medium alone) ranged from 210–321. ∗, p < 0.05 vs cultures without gamma-irradiated MSC.

Close modal

This study reports the immunological properties of MSC and the implication for the use of these cells in repair medicine. The phenotype of the cells is consistent for MSC (7). Furthermore, MSC can undergo adipogenic and osteogenic differentiation (Fig. 1, D and E), suggesting that the cells are indeed MSC. However, since cell cloning was not performed, it is possible that the MSC used in the study are heterogeneous. We showed that MSC can elicit allogeneic responses (Fig. 3) in one-way MLR, but not as third-party cells in MLR. However, MSC possess the ability to mediate immunosuppression (Figs. 5,A and 6,A). Interestingly, MSC express MHC class II in a distinct fashion, showing high expression close to the nuclei in a punctated manner (Fig. 2,B). Flow cytometric analysis for MHC class II displayed a broad peak, suggesting that MHC class II expression is variable in individual MSC (Fig. 2,A). For reactions in which MSC induced proliferation of PBMC, the soluble factors produced by MSC (Table II) did not seem to be important in mediating the allogeneic effects of MSC (Fig. 4,A). Detection of several broad families of growth factors in the supernatants of confluent MSC (Table II) suggested that in the absence of an immune challenge, MSC might be producing a balanced group of soluble factors to maintain homeostasis. Interestingly, MSC did not suppress the proliferative responses to recall Ags (Fig. 7). This discriminatory property of MSC makes these cells clinically useful, since they might not deter normal immune properties of MSC when used in tissue repair. The intriguing aspect of this study is the veto-like effects of MSC (Fig. 6). This property is not important for the use of these cells in trans-differentiation studies, but perhaps MSC might also be used in organ and BM transplantations to inhibit alloantigen-induced host responses.

Although there are several reports on the isolation of MSC (37), in our hands, we have standardized the most efficient method to obtain highly purified MSC. The study was pursued with the intent that MSC could trans-differentiate into cells of different tissues (26). The issue regarding the ability of MSC to elicit allogeneic responses is unclear and has been addressed in detail in this study. At present, it is unclear how MSC would be applied in repair medicine, e.g., should MSC be transplanted as unmanipulated cells or should cells be implanted as fully or partly differentiated? However, regardless of their stage of maturation, the allogeneic properties of MSC would be relevant.

The allo-suppressive effect of MSC is specific since similar suppressive effects were not observed for BM fibroblasts from the same donor (Fig. 5,B). The BM fibroblasts used in the studies were pure, with no detectable contaminating cells such as macrophages (not shown). BM fibroblasts represent the major subset of stromal cells and can be differentiated from MSC (38, 39). Thus, our findings differ from other recent studies by Di Nicola et al. (40), which showed immunosuppressive effects of BM fibroblasts. The contrast between studies by Di Nicola (40) and those in our report could be explained by differences in the experimental design. Studies by Di Nicola et al. (40) used purified T cells from peripheral blood as their responding population, while we used the entire mononuclear cell fraction. Our goal was to establish a condition that could model a situation of graft-vs-host-like responses. Our results pertaining to cytokine production are consistent with those reported by Di Nicola (40), in which a role for TGF-β was established (Table II). Further studies regarding the data on cytokines (Table II) are needed, and extrapolation of the findings would be speculation.

The immune-suppressive functions of MSC are supported by two independent reports (27, 40). Noort et al. (41) showed that MSC promote engraftment of human cells in SCID mice. This property of MSC in engraftment could not be explained by MSC forming their own stromal cells, since longitudinal studies with human transplant patients showed that the stromal cells were of host origin (42). The facilitating effects of MSC in BM transplantation might be explained by the veto-like functions of MSC. Thus, contrary to current thought that MSC could facilitate engraftment by forming stromal layers, the same end point might be explained by another mechanism, suppression of graft-vs-host disease during transplantation. The studies presented in this report do not discriminate concerning whether MSC exert suppressive functions in both direct and indirect pathways by alloantigen (43). Further experimental studies are warranted to address this pertinent question to understand the immunology of MSC. Further basic and clinical research into the biology of MSC in the BM and identification of a possible precursor of MSC will facilitate their use in repair medicine.

1

This work was supported by National Institutes of Health Grants HL54973 and CA89868 and The F. M. Kirby Foundation. It was submitted for publication in partial fulfillment of the requirements for the Ph.D. degree from the Graduate School of Biomedical Sciences, University of Medicine and Dentistry of New Jersey (J.A.P.).

3

Abbreviations used in this paper: BM, bone marrow; LHSC, lymphohemopoietic cells; MSC, mesenchymal stem cells; PFA, paraformaldehyde; S.I., stimulation index.

1
Zon, L. I..
1995
. Developmental biology of hematopoiesis.
Blood
86
:
2876
.
2
Deans, R. J., A. B. Moseley.
2000
. Mesenchymal stem cells: biology and potential clinical uses.
Exp Hematol.
28
:
875
.
3
Jiang, Y., B. N. Jahagirdar, R. L. Reinhardt, R. E. Schwartz, C. D. Keene, R. Ortiz-Gonzalez, M. Reyes, T. Lenvik, T. Lund, M. Blackstad, et al
2002
. Pluripotency of mesenchymal stem cells derived from adult marrow.
Nature
418
:
41
.
4
Arai, F., O. Ohneda, T. Miyamoto, X. Q. Zhang, T. Suda.
2002
. Mesenchymal stem cells in perichondrium express activated leukocyte cell adhesion molecule and participate in bone marrow formation.
J. Exp. Med.
195
:
1549
.
5
Bianco, P., M. Riminucci, S. Gronthos, P. G. Robey.
2001
. Bone marrow stromal stem cells: nature, biology, and potential applications.
Stem Cells
19
:
180
.
6
Campagnoli, C., I. A. G. Roberts, S. Kumar, P. R. Bennett, I. Bellantuono, N. M. Fisk.
2002
. Identification of mesenchymal stem/progenitor cells in human first-trimester fetal blood, liver, and bone marrow.
Blood
98
:
2396
.
7
Pittenger, M. F., A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak.
1999
. Multilineage potential of adult human mesenchymal stem cells.
Science
284
:
143
.
8
Akashi, K., D. Traver, T. Miyamotot, I. L. Weissman.
2000
. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
Nature
404
:
193
.
9
Chow, D. C., L. A. Wenning, W. M. Miller, E. T. Papoutsakis.
2001
. Modeling pO2 distributions in the bone marrow hematopoietic compartment. I. Krogh’s model.
Biophys. J.
81
:
675
.
10
Chow, D. C., L. A. Wenning, W. M. Miller, E. T. Papoutsakis.
2001
. Modeling pO2 distributions in the bone marrow hematopoietic compartment. II. Modified Kroghian models.
Biophys. J.
81
:
685
.
11
Harrison, J. S., P. Rameshwar, V. Chang, P. Bandari.
2002
. Oxygen saturation in the bone marrow of healthy volunteers.
Blood
99
:
394
.
12
Hawley, R. G., D. A. Sobieski.
2002
. Somatic stem cell plasticity: to be or not to be.
Stem Cells
20
:
195
.
13
D’Amour, K. A., F. H. Gage.
2002
. Are somatic stem cells pluripotent or lineage-restricted?.
Nat. Med.
8
:
213
.
14
Schwartz, R. E., M. Reyes, L. Koodie, Y. Jiang, M. Blackstad, T. Lund, T. Lenvik, S. Johnson, W. S. Hu, C. M. Verfaille.
2002
. Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells.
J. Clin. Invest.
109
:
1291
.
15
Jin, H. K., J. E. Carter, G. W. Huntley, E. H. Schuchman.
2002
. Intracerebral transplantation of mesenchymal stem cells into acid sphingomyelinase-deficient mice delays the onset of neurological abnormalities and extends their life span.
J. Clin. Invest.
109
:
1183
.
16
Devine, S. M., A. M. Bartholomew, N. Mahmud, M. Nelson, S. Patil, W. Hardy, C. Sturgeon, T. Hewett, T. Chung, W. Stock, et al
2001
. Mesenchymal stem cells are capable of homing to the bone marrow of non-human primates following systemic infusion.
Exp. Hematol.
29
:
244
.
17
Lemischka, I..
2002
. A few thoughts about plasticity of stem cells.
Exp. Hematol.
30
:
848
.
18
Sanchez-Ramos, J., S. Song, F. Cardozo-Pelaez, C Hazzi, T. Stedeford, A. Willing, T. B. Freeman, S. Saporta, W. Janssen, N. Patel, et al
2000
. Adult bone marrow stromal cells differentiate into neural cells in vitro.
Exp. Neurol.
164
:
247
.
19
Woodbury, D., E. Schwarz, D. Prockop, I. Black.
2000
. Adult rat and human bone marrow stromal cells differentiate into neurons.
J. Neurosci. Res.
61
:
364
.
20
Kopen, G. C., D. J. Prockop, D. G. Phinney.
1999
. Marrow stromal cells migrate through forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains.
Proc. Natl. Acad. Sci. USA
96
:
10711
.
21
Azizi, S. A., D. Stokes, B. J. Augelli, C. DiGirolamo, D. J. Prockop.
1998
. Engraftment and migration of human bone marrow stromal cells implanted in the brains of albino rats: similarities to astrocyte grafts.
Proc. Natl. Acad. Sci. USA
95
:
3908
.
22
Wakitani, S., T. Saito, A. I. Caplan.
1995
. Myogenic cells derived from rat bone marrow mesenchymal stem cells exposed to 5-azacytidine.
Muscle Nerve
18
:
1417
.
23
Gerson, S. L..
1999
. Mesenchymal stem cells: no longer second class marrow citizens.
Nat. Med.
5
:
262
.
24
Caplan, A. I., S. R. Bruder.
2001
. Mesenchymal stem cells: building blocks for molecular medicine in the 21st century.
Trends Mol. Med.
7
:
259
.
25
Botella, L. M., T. Sanchez-Elsner, F. Sanz-Rodriguez, S. Kojima, J. Shimada, M. Guerrero-Esteo, M. P. Cooreman, V. Ratziu, C. Langa, C. P. H Vary, et al
2002
. Transcriptional activation of endoglin and transforming growth factor-β signaling components by cooperative interaction between Sp1 and KLF6: their potential role in the response to vascular injury.
Blood
100
:
4001
.
26
Neumann, H..
2001
. Control of glial immune function by neurons.
Glia
36
:
191
.
27
Bartholomew, A., C. Sturgeon, M. Siatskas, K. Ferrer, K. McIntosh, S. Patil, S., W. Hardy, S. Devine, D. Ucker, R. Deans, et al
2002
. Mesenchymal stem cells suppress lymphocyte proliferation in vitro and prolong skin graft survival in vivo.
Exp. Hematol.
30
:
42
.
28
Thomas, J. M., F. M. Carver, P. R. Cunningham, L. C. Olson, F. T. Thomas.
1991
. Kidney allograft tolerance in primates without chronic immunosuppression: the role of veto cells.
Transplantation
51
:
198
.
29
Miller, R. G..
1980
. An immunological suppressor cell inactivating cytotoxic T-lymphocyte precursor cells recognizing it.
Nature
287
:
544
.
30
Nakamura, H., R. E. Gress.
1990
. Interleukin 2 enhancement of veto suppressor cell function in T-cell-depleted bone marrow in vitro and in vivo.
Transplantation
49
:
931
.
31
Asiedu, C., Y. Meng, W. Wang, Z Huang, J. L. Contreras, J. F. George, J. M. Thomas.
1999
. Immunoregulatory role of CD8α in the veto effect.
Transplantation
67
:
372
.
32
Haynesworth, S. E., M. A. Baber, A. I. Caplan.
1992
. Cell surface antigens on human marrow-derived mesenchymal cells are detected by monoclonal antibodies.
Bone
13
:
69
.
33
Rameshwar, P., A. Poddar, G. Zhu, P. Gascon.
1997
. Receptor induction regulates the synergistic effects of substance P with IL-1 and PDGF on the proliferation of bone marrow fibroblasts.
J. Immunol.
158
:
3417
.
34
Majumdar, M. K., M. A. Thiede, J. D. Mosca, M. Moorman, S. L. Gerson.
1998
. Phenotypic and Phenotypic and functional comparison of cultures of marrow-derived mesenchymal stem cells (MSCs) and stromal cells.
J. Cell. Physiol.
176
:
57
.
35
Singh, D., D. D. Joshi, M. Hameed, J. Qian, P. Gascon, P. B. Maloof, A. Mosenthal, P. Rameshwar.
2000
. Increased expression of preprotachykinin-I and neurokinin receptors in human breast cancer cells: implications for bone marrow metastasis.
Proc. Natl. Acad. Sci. USA
97
:
388
.
36
Reyes, M, T. Lund, T. Lenvik, D. Aguiar, L. Koodie, C. M. Verfaillie.
2001
. Purification and ex vivo expansion of postnatal human marrow mesodermal progenitor cells.
Blood
98
:
2615
.
37
Conrad, C., B. Gottgens, S. Kinston, J. Ellwart, R. Huss.
2002
. GATA transcription in a small rhodamine 123lowCD34+ subpopulation of a peripheral blood-derived CD34CD105+ mesenchymal cell line.
Exp. Hematol.
30
:
887
.
38
Minguell, J. J., A. Erices, P. Conget.
2001
. Mesenchymal stem cells.
Exp. Biol. Med.
226
:
507
.
39
Dennis, J. E., P. Charbord.
2002
. Origin and differentiation of human and murine stroma.
Stem Cells
20
:
205
.
40
Di Nicola, M., C. Carlo-Stella, M. Magni, M. Milanesi, P. D. Longoni, P. Matteucci, S. Grisanti, A. M. Gianni.
2002
. Human bone marrow stromal cells suppress T-lymphocyte proliferation induced by cellular or nonspecific mitogenic stimuli.
Blood
99
:
3838
.
41
Noort, W. A., A. B. Kruisselbrink, P. S. Anker, M. Kruger, R. L. van Bezooijen, R. A. de Paus, M. H. Heemskerk, C. W. Lowik, J. H. Falkenburg, R. Willemze, et al
2002
. Mesenchymal stem cells promote engraftment of human umbilical cord blood-derived CD34+ cells in NOD/SCID mice.
Exp. Hematol.
30
:
870
.
42
Awaya, N., K. Rupert, E. Bryan, B. Torok-Storb.
2002
. Failure of adult marrow-derived stem cells to generate marrow stroma after successful hematopoietic stem cell transplantation.
Exp. Hematol.
30
:
937
.
43
Haller, G. W., B. Lima, S. M. Kunisaki, S. Germana, C. Leguern, C. A. Huang, D. H. Sachs.
2002
. MHC alloantigens elicit secondary, but not primary, indirect in vitro proliferative responses.
J. Immunol.
169
:
3613
.