Experimental Cell Research 295 (2004) 395 406 [Link].

com/locate/yexcr

Isolation and characterisation of mesenchymal stem cells from adult mouse bone marrow
` Philippe Tropel, a,* Daniele Noel, b Nadine Platet, a Pierre Legrand, c Alim-Louis Benabid, a and Francois Berger a
INSERM U318, Grenoble, France INSERM U475, Montpellier, France c EMBL, Grenoble-outstation, Grenoble, France
b a

Received 19 August 2003, revised version received 2 December 2003

Abstract The future use of adult mesenchymal stem cells (MSCs) for human therapies depends on the establishment of preclinical studies with other mammals such as mouse. Surprisingly, purification and characterisation of murine MSCs were only poorly documented. The aim of this study was to purify mouse MSCs from adult bone marrow and to functionally characterise their abilities to differentiate along diverse lineages. Adherent cells from adult C57Bl/6J mouse bone marrow were depleted of granulo-monocytic cells and subsequently allowed to grow on fibronectin-coated dishes in presence of fetal bovine serum and growth factors. The growing fibroblastoid cell population primarily consisted of spindle- and star-shaped cells with significant renewal capacity as they were cultured until 30 passages (about 60 doubling population). We fully demonstrated the MSC phenotype of these cells by inducing them to differentiate along osteoblastic, adipocytic, and chondrocytic pathways. Mouse MSCs (mMSCs) sharing the same morphological and functional characteristics as human MSCs can be successfully isolated from adult bone marrow without previous mouse or bone marrow treatment. Therefore, mMSCs will be an important tool to study the in vivo behaviour and fate of this cell type after grafting in mouse pathology models. D 2004 Elsevier Inc. All rights reserved.
Keywords: Mesenchymal stem cell; Mouse; Bone marrow

Introduction Stem cells were initially defined as cellular entities with two main properties: self-renewal (producing daughter stem cells with identical potentialities) and the ability to differentiate along one or more lineages [1]. Current models describe two stem cell populations with distinct progenies that are housed within adult bone marrow, hematopoietic stem cells (HSCs) and mesenchymal stem cells (MSCs). All lymphoid and myeloid lineages that finally produce blood circulating cells and organ-resident cells of the immune response originate from HSCs [2,3]. HSCs renewal and differentiation steps are tightly regulated by the bone marrow microenvironment, which is composed of osteo-

* Corresponding author. CHU de Grenoble, INSERM U318, Pavillon B, BP 217, 38043 Grenoble Cedex 9, France. Fax: +33-4-76-76-56-19. E-mail address: ptropel@[Link] (P. Tropel). 0014-4827/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/[Link].2003.12.030

blasts, adipocytes, endothelial cells and vascular pericytes [4]. Except for endothelial cells whose origin may be multiple [5,6], cells of the microenvironment may derive from a unique pluripotent stem cell, the MSC, also called a stromal stem cell [7 9]. MSCs gold standard was defined in human by Pittenger et al. [9] as bone marrow-derived fibroblasts able to differentiate under appropriate stimuli along three principal lineages: osteoblastic, adipocytic and chondrocytic lineages. This functional definition allows affirmation of the MSC nature of a cell population in the absence of strict specific markers. Nevertheless, although specific cell surface markers are lacking to strictly identified human MSCs (unpublished results cited in Ref. [10]), enrichment strategies were developed based on selection of cells positive for the NGF receptor [11], STRO-1 antibody-recognised antigen [12 14] and the SH-2, -3 and -4 antibodies recognised antigens [15,16], or based on cell sorting by size selection [17]. However, these approaches are still marginal. MSC

396

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

purification is actually currently achieved by a single step of adhesion to cell culture plastic, a unique property among human bone marrow cells [9,18,19]. Alternatively, Verfaillies team reported in 2001 the purification of a fibroblastoid cell population from human, rat and mouse bone marrow [20,21], called multipotent adult progenitor cells (MAPCs), with enlarged differentiation possibilities, but with an unclear relationship to MSC. Actually, it might be either a more primitive or a completely different stem cell population with partially overlapping ontogeny. MSCs represent an ideal stem cell source for cell therapies because of their easy purification and amplification, and their multipotency. As an example, clinical assays were initiated in the last years showing MSC-induced improvement of bone growth and mineralisation in children affected with osteogenesis imperfecta [22,23], a genetic disorder characterised by defective type I collagen. Nevertheless, MSC multipotency suggests that these cells may be used to cure other human pathologies [24,25]. This point has justified the development of animal models for preclinical studies. To this aim, the isolation of MSCs from the bone marrow of rat [26], cat [27], dog [28], baboon [29], rabbit [30,31], pig [32], goat [33] and sheep [34] were successfully completed. In contrast, mouse MSC purification was only poorly reported despite their potential use in both physiology and pathophysiological mammal models. Many MSC-like immortalised cell lines with large proliferation and differentiation potentialities were obtained from mouse bone marrow [35 39] but their immortalised status precluded all conclusion about MSC physiologic function and physiology. In another way, mouse mesenchymal progenitors could be isolated from mouse bone marrow using a sophisticated procedure involving cell transduction with a selectable marker-carrying vector [40]. Alternatively, the authors used unpurified adherent culture of mouse bone marrow [41,42], which was shown to be contaminated by hematopoietic cells (Ref. [43] and our unpublished results). It clearly appeared that primary mesenchymal stem cells derived from adult mouse bone marrow are lacking to study MSC physiology or to test grafting in multiple disposable genetic mouse models. Therefore, we initiated work to purify mouse MSCs and fully characterise their differentiation potentialities along mesenchymal pathways. A previously published protocol [44] was modified for this purification, based on immunodepletion of CD11b-positive monocytic cells from the bone marrow culture. This protocol did not preclude the subsequent development of hematopoietic colonies in secondary culture, but finally allowed a fibroblastoid cell population to grow to a sustainable level after several weeks. To demonstrate that the population we purified was MSC in nature, we first stimulated these cells to induce their differentiation into the three different lineages described by Pittenger: osteoblastogenesis, adipogenesis and chondrogenesis. We then demonstrated the expression of various markers specific to these pathways.

Materials and methods Mesenchymal stem cell isolation and culture Mouse stromal cells were isolated according to a protocol modified from Kopen et al. [44]. Bone marrow was collected from 2-month-old C57Bl/6 mice (Charles River, LArbresles, France) by flushing femurs and tibias with complete medium constituted of DMEM-LG (BioWhittaker, Verviers, Belgium), 10% fetal calf serum (FCS, Sigma, St. Louis, USA.), glutamine 2 mM (BioWhittaker) and penicillin/streptomycin (50 U/ml and 50 mg/ml, respectively; Gibco-Invitrogen, Carlsbad, USA) supplemented with heparin at a final concentration of 5 U/ml. Cells were then washed twice in a medium without heparin and plated in a Petri dish at a density of 2 106 cells/cm2. After 3 days, nonadherent cells were removed by two to three washes with PBS 1 and adherent cells further cultured in complete medium for 4 supplementary days. At this time, adherent cells were retrieved by trypsinisation and immunodepleted of granulo-monocytic cells using a biotinylated antibody against CD11b (BD Pharmingen, Pont-de-Claix, France), and streptavidine-coated microbeads from Miltenyi Biotec (Paris, France) according to the manufacturers instructions. CD11b-negative cells were subsequently seeded on human fibronectin (Fn, BD Pharmingen)-coated Petri dishes at a final density of 1000 cells/cm2 in complete medium supplemented with EGF 10 ng/ml and PDGFAA 10 ng/ml (from R&D Systems, Minneapolis, USA) until outbreak of a fibroblastoid cell population (between 1 and 2 months after bone marrow plating). Fibroblastoid cells were routinely cultured and amplified on Fncoated Petri dishes in complete medium without growth factors. Medium was changed two to three times a week and cell density was maintained between 2000 and 15,000 cells/cm2. Viral transduction Mouse MSCs were infected with Moloney Mouse Leukemia Virus-derived eGFP carrying retrovirus (modified from Ref. [45]) kindly provided by Dr A. Cumano (Institut Pasteur, Paris, France). Practically, cells were cultured for 2 days in the presence of an eGFP carrying Moloney Mouse Leukemia Virus-derived vector supernatant supplemented with polybrene at 5 Ag/ml. Cells were rinsed twice and then amplified. Transduced cells were analysed by FACS for eGFP fluorescence. Tumorigenic assay Two-month-old SWISS-nude mice (n = 3; Charles River Laboratories, LArbresle, France) were subcutaneously injected with 5 105 cells in 50 Al of complete medium. In the case of purified mMSCs, SWISS-nude

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

397

mice were followed for 6 months. Positive control consisted of GL26 glioma cells [46] cultured and injected in the same conditions. Fluorescence-activated cell sorting analysis of hematopoietic markers Mesenchymal stem cells and whole bone marrow were retrieved as described above. As MSCs were retrieved by culture trypsinisation, bone marrow was submitted to the same treatment before cell labelling. Preliminary experiments have shown that trypsin treatment did not affect the number of labelled cells and lightly increased fluorescence mean, suggesting that enzymatic reaction improved antigen accessibility. Trypsin was neutralised by the addition of PBS 1 containing 5% heat-inactivated FBS and 0.16% sodium azide (Sigma). Cells were washed twice and resuspended at 3 5 106 cells/ml in the buffer containing biotinylated antibody diluted at 1 Ag/ml. Anti-CD11b and anti-CD45 were purchased from BD Pharmingen. Labelling reaction was shaked for 10 min at room temperature. Cells were washed twice and then labelled with ExtrAvidin-FITC diluted to 1:400 (Sigma). After two more washes, FACS analysis was performed on a FACScalibur (BD) flow cytometer using CellQuestk software with 20,000 events recorded for each sample. Immunofluorescence The medium was removed and cells were washed twice with PBS before fixation in 4% paraformaldehyde solution in PBS for 10 min. Cells were rinsed three times with PBS and incubated in a saturation/permeabilisation PBS solution containing Triton X-100 0.3% and bovine serum albumin

(BSA) 4% for 20 min. Solution was removed and replaced by anti-CD11b biotinylated antibody (BD Biosciences Pharmingen, San Diego, USA) at 5 Ag/ml in PBS BSA 3% for 30 min. After three PBS washes, slides were incubated for 30 min with a TRITC-labelled ExtrAvidin (Sigma) solution (1/200) in the same buffer used for primary antibody solution. Genomic DNA was then labelled with a DAPI solution in PBS for 15 additional minutes before final washes and mounting in FluorSave (Calbiochem, San Diego, USA). Differentiation of mouse mesenchymal stem cells Osteoblastic differentiation was induced by culturing confluent mouse MSCs for 3 weeks in complete medium supplemented with dexamethasone 108 M, h-glycerophosphate 10 mM, and ascorbic acid 0.3 mM (all from Sigma). Cells were maintained in inducing medium for 3 weeks with daily change to overcome the instability of ascorbic acid in neutral pH. Then, cells were fixed for 2 min in ice-cold methanol before labelling for alkaline phosphatase activity according to Reyess protocol [20]. Alternatively, treated cells were used for RNA extraction and RT-PCR analysis of gene expression. To induce adipocytic differentiation, confluent MSCs were cultured from 1 to 3 weeks in DMEM-HG 40%, HamF12 50% (both from BioWittaker), rabbit serum 10% (Sigma), dexamethasone 108 M, insulin 0.5 Ag/ml (Sigma), glutamine 2 mM and antibiotics. Cells were used for RNA extraction or for lipid droplet staining using S Red Oil (Sigma) [20]. To induce chondrocytic differentiation, harvested MSCs were rinsed once by centrifugation, resuspended in chondrogenic medium, and finally centrifuged 5 min at 350 g

Table 1 Specific primers for RT-PCR amplification Marker C/EBPa PPARg1 PPARg2 Lipoprotein lipase aP2 Adipsin CBFA1 Type I collagen Osteopontin Osteonectin Osteocalcin Type II collagen Type XI collagen Aggrecan HPRT GAPDH Forward GCCGCCTTCAACGACGAG TTCTGACAGGACTGTGTGACAG GCTGTTATGGGTGAAACTCTG GAAGGGAAAGGACTCAGCAG GGGATTTGGTCACCATCCG CTGCTGGACGAGCAGTGG CCGCACGACAACCGCACCAT GAAGTCAGCTGCATACAC TCACCATTCGGATGAGTCTG AGCGCCTGGAGGCTGGAGAC TCTGCTCACTCTGCTGAC GCCTCGCGGTGAGCCTGATC GGAAAGATGGGCTACCAGGACA CCAAGTTCCAGGGTCACTGTTACCG GCTGGTGAAAAGGACCTCT TGAAGGTCGGTGTGAACGGATTTGGC Reverse TGGCCAGGCTGTAGGTGCAT ATAAGGTGGAGATGCAGGTTC Same reverse as PPARg1 ATCAGAAGACATCAGGCAGG CCAGCTTGTCACCATCTCG GATGACACTCGGGTATAGACGC CGCTCCGGCCCACAAATCTC AGGAAGTCCAGGCTGTCC ACTTGTGGCTCTGATGTTCC CTTGATGCCAAAGCAGCCGG GGAGCTGCTGTGACATCC CTCCATCTCTGCACGGGGT GGACGTCCTGGCAAACCAATTG TCCTCTCCGGTGGCAAAGAAGTTG CACAGGACTAGAACACCTGC CATGTAGGCCATGAGGTCCACCAC Size 449 354 351 353 204 569 289 313 437 269 388 IIA: 472 IIB: 268 412 271 250 983 Reference [79] [80] [80] HM [79] [81] [71] [82] [83] HM [82] HM [84] HM [48] HM

Following primer sequences, sizes (in base pairs) of amplified fragments and references are indicated. HM: homemade. Two alternatively spliced mRNA variants, IIA and IIB, transcribed from the type II collagen encoding gene can be analysed using our specific primers. Respective sizes of each amplified fragment are given.

398

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

in 15 ml polypropylene Falcon tube to form a pellet. Chondrogenic medium was constituted of DMEM-HG, dexamethasone 107 M, sodium pyruvate 1 mM, 2-phosphate ascorbic acid 1.7 104 M, proline 35 mM, ITS + 2 (all products from Sigma) and antibiotics and recovered after 48 h culture on BMP-2-expressing C9 cells [47]. Medium was changed every 3 4 days for 3 weeks before RNA extraction or alcian blue (Sigma) staining of cartilage proteoglycans after paraffin embedding. RNA extraction and RT-PCR analysis of gene expression All RNA extractions were performed with RNAgents kit from Promega (Madison, USA). Cells were washed twice with PBS before adding lysis buffer. Reverse transcriptions and PCR amplification of cDNAs were done as previously described [48]. Primer sequences are given in Table 1. X-ray diffraction of mineralised extracellular matrix After cell lysis for RNA extraction, plates were extensively washed with deionised water. Mineralised matrixes were retrieved in water by mechanical scraping using a scalpel blade. Matrixes were further rinsed with water, desiccated and finally mechanically powdered. The powder was mounted on very thin nylon loops. The X-ray powder diffraction images were recorded on an MAR345 imageplate detector mounted on a M18XHF copper rotating anode generator (Siemens, Madison, USA) equipped with focussing mirrors (Rigaku/MSC, Sevenoaks, United Kingdom) and running at 40 kV and 100 mA. Reference diffraction images were recorded using hydroxyapatite (Sigma).

Fig. 1. Mouse mesenchymal stem cell isolation. CD11b-negative cells were plated on fibronectin in complete medium supplemented with EGF and PDGF (A). After 1 week, widely spread cells were seen tightly interacting with smaller cells (B), which were further identified such as granulomonocytic CD11b-positive cells by immunofluorescence (C). Arrowheads in B and C indicate the nuclei of stromal cells. In C, all CD11b-positive cells were seen interacting with the stromal cell cytoplasm in a similar manner than in B. A and B are May Grunwald Giemsa staining (original magnification 100). C is originally magnified 100.

Results Establishment of a mesenchymal stem cell population from mouse bone marrow MSC isolation from human bone marrow is classically performed by adhesion to plastic. However, this technique is not suitable for mouse MSC purification due to the unwanted growth of granulo-monocytic cells in both primary and passaged cultures [43,49,50]. To overcome this problem, adherent cell population was retrieved from bone marrow culture, immunodepleted of granulo-monocytic cells using an anti-CD11b specific antibody [44] and subcultured. CD11b-negative cells showed widespread morphologies with fibrous cytoplasms (Fig. 1A) that we hypothesised to be previously described myofibroblasts [7]. Very few fibroblasts were visible. About 1 week after this first step, a new cell population appeared and became predominant after approximately 2 weeks. These cells closely interacted with myofibroblasts (Fig. 1B) and were positive for CD11b marker indicating they belonged to the granulo-monocytic lineage (Fig. 1C). This hematopoietic

population continued to be present for many weeks until a new population with a fibroblastoid shape appeared and progressively became the predominant cell type present. After 10 passages, we tested the presence of hematopoietic cells by FACS analysis, using anti-CD45 (Fig. 2A) and CD11b (Fig. 2B) specific antibodies, expressed by hematopoietic cells and granulo-monocytic cells, respectively. FACS results showed that there was no contaminating hematopoietic population. Moreover, fibroblastic cells expressed neither CD45 (Fig. 2A) nor CD11b (Fig. 2B). These data demonstrate that the culture was devoid of contaminating hematopoietic cells. Hematopoietic cell clearance was probably due to progressive loss of progenitors. Fibroblast cell population shared diverse morphologies: spindle-shaped, star-shaped and myofibroblastoid cells (Fig. 2C). The doubling time was 1.7 F 0.4 days (mean F SEM). The culture was expanded for more than 30 passages representing a minimum of 60 doublings. To insure that the purified cell population did not contain transformed cells with abnormal proliferative characteristics, we first cultured it without serum. In this condition, cells showed senescence signs and died in few days (P.T., data not shown). We then tested their in vivo tumorigenic potentialities in immunodeficient mice. When GL26 glioma cells (5 105 cells) were subcutaneously injected in SWISS-nude mice, all animals developed a tumour within 2 weeks. Then, 5 105 bone marrow-derived fibroblastic cells were injected in a same way (n = 3) with no localised tumor formation after 6 months (P.T., data not shown). In summary, these data demonstrated that our fibroblastoid cell population was free of transformed or tumorigenic cells.

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

399

Fig. 2. Hematopoietic cell-free culture of mouse mesenchymal stem cells. Fibroblastic cell cultures (passage 10) were analysed by FACS for the expression of hematopoietic markers, CD45 (A) and CD11b (B). Whole bone marrow was freshly retrieved from adult C57Bl/6J mice, trypsinised and analysed in the same way as a positive control; BM: bone marrow; mMSCs: mouse mesenchymal stem cells; blue line: control cells labelled without primary antibodies; red line: cells labelled with primary antibodies and ExtrAvidin-FITC. The mMSC culture was shown to be completely devoid of CD11b- and CD45-positive hematopoietic cells. In C, the fibroblastoid morphology of mMSC is shown (original magnification 200; phase contrast microscopy).

Functional characterisation As defined by Pittenger et al. [9], we wanted to determine whether MSCs are able to differentiate along three lineages: osteoblastic, adipocytic and chondrocytic. Differentiation along the osteoblastic lineage and accumulation of a mineralised bone-like matrix To induce osteoblastic differentiation, confluent cells were further cultured in osteoinductive medium consisting of complete medium supplemented with h-glycerophosphate, dexamethasone and ascorbic acid [51]. We first analysed the alkaline phosphatase activity as it is considered a marker of osteoblastic differentiation. In the initial culture, alkaline phosphatase positive cells were very rare (Figs.

3A,B). After 3 weeks in inductive medium, the treated population contained numerous positive cells widely distributed in the dish and others grouped in colonies (Figs. 3C,D). Nevertheless, many different cell types can express alkaline phosphatase activity. To further demonstrate the differentiation of purified cells, we analysed by RT-PCR the expression of osteoblast-specific markers before and after treatment with the inductive medium (Fig. 3E). Mouse MSCs maintained under basal conditions constitutively expressed certain osteoblastic lineage markers. CBFA1/ Runx2, type I collagen a chain, osteopontin and osteonectin mRNAs were all easily detectable in this culture (Fig. 3E). In contrast, osteocalcin, a marker for advanced osteoblastic differentiation, was barely detectable. After the third week of culture in osteoinductive medium, CBFA1, type I colla-

400

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

deposits near the center of alkaline phosphatase positive colonies (Fig. 4A). As mineralisation can be due to nonspecific processes, we formally characterised the accumulated mineral. Mineralised matrix was retrieved after differentiation and characterised by X-ray diffraction analysis. The X-ray diffraction pattern showed major rings in the 30 35j2H zone, very similar to that produced by the hydroxyapatite spectrum (Fig. 4B). This data strongly supported a bone-type mineralisation of the extracellular matrix secreted by differentiated mouse MSCs. Differentiation along the adipocytic lineage To induce adipocytic differentiation, we selected a rabbit serum, dexamethasone and insulin-containing medium to obtain a massive lipid droplet accumulation illustrated by O Red oil positive staining (Fig. 5A). Lipid droplets were detectable after only 4 days, but 2 weeks were necessary to accomplish a maximal lipid accumulation. Differentiation was further demonstrated by RT-PCR analysis of adipocytic markers expression. Mouse MSCs in basic culture conditions expressed lipoprotein lipase, aP2, and transcription factors known to be involved in control of adipocytic differentiation such as C/EBPa and PPARg1 (Fig. 5B). After a 3-week induction period, cells expressed two supplementary markers of mature adipocytes, PPARg2 (an alternatively spliced isoform of PPARg) and adipsin. A longer induction period led to adipocyte loss by detachment from the matrix. Differentiation along the chondrocytic lineage We ultimately tested the chondrogenic potentialities of the cell population when cultured in micropellet. Cells

Fig. 3. Osteoblastic differentiation. In a first approach, alkaline phosphatase activity was tested as an osteoblastic differentiation marker. A and C are photographs of stained plates before and after osteoinductive treatment, respectively. B and D are higher magnification of representative regions of A and C. The untreated fibroblastoid cell culture was found negative for alkaline phosphatase activity (A, B). In contrast, cells treated for 3 weeks with an ascorbic acid, dexamethasone and h-glycerophosphate supplemented medium revealed a strong positive signal (C, D) (brown black staining, methylene blue counterstain, original magnification 200). (E) Expression of osteoblast marker encoding genes was analysed by RT-PCR before (D0) and after differentiation culture (D20). Pictures are BET-stained electrophoresis gels of PCR products. HPRT is shown as a control for RNA sample quality.

gen, osteopontin and osteonectin mRNA levels were unchanged or slightly increased. In contrast, osteocalcin mRNA was strongly induced. Parallel to the differentiation process, mineral accumulation occurred progressively during the treatment period. Mineral deposition was widely distributed in the culture dish, as were numerous foci of

Fig. 4. Apatite accumulation during osteoblastic differentiation of mouse mesenchymal stem cells. Osteoblast differentiated cells showed mineral accumulation easily viewed such as a white deposition both widely distributed in the dish and in the center of colonies. In A, both types of accumulation are shown. White arrows: external border of an alkaline phosphatase labelled osteoblastic colony (black staining). Black arrowheads: limits of the white mineral deposition accumulated in the center of the colony during its growth. Mineral matrix powders were subjected to an X-ray beam, and diffraction patterns were recorded and compared to the hydroxyapatite reference. (B) I, II and III: diffraction patterns of three different mineralised matrix samples. IV: diffraction pattern of standard hydroxyapatite. Diffraction rings position: (1) 10.8j2H; (2) 24.9j2H; (3) 28.8j2H; (4) 31.96j2H.

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

401

predominantly the type IIB collagen, aggrecan, another marker for mature chondrocytes, and high levels of the type XI collagen, demonstrating that cells have undergone chondrocytic differentiation. Efficient transduction of mouse MSCs by eGFP-carrying retroviral vector We finally demonstrated that mMSCs could be genetically modified using classical retroviral transduction procedures. Cells were cultured for 2 days in the presence of a medium conditioned with an eGFP MMLV-derived vector, and then amplified and frozen. Transduction efficiency was evaluated by FACS analysis of eGFP fluorescence. Fortyone to eighty percent of cells were found expressing eGFP (Fig. 7). EGFP expressing fraction and mean of expression per cell were not changed after two freezings and four further passages. Neither viral infection nor eGFP expres-

Fig. 5. Adipocytic differentiation. Confluent isolated cells were cultured for 2 3 weeks in dexathemasone, insulin and rabbit serum containing medium. Lipid droplet could be seen as early as 5 days after induction. After 2 weeks, lipid accumulation was clearly visible and stained with O Red Oil (A, methylene blue couterstaining, original magnification 100). (B) RTPCR analysis of adipocyte specific markers expression before (D0) and after differentiation (D20). Pictures are BET-stained electrophoresis gels of PCR products. HPRT is shown as a control for RNA sample quality.

were spun and allowed to aggregate favouring closed cell interactions and 3-dimensional constraints. Aggregates were cultured as free-floating in a serum-free medium supplemented with BMP-2. After 3 weeks, the micropellet were positively stained with alcian blue, suggesting an accumulation of cartilaginous proteoglycans (Fig. 6A). RTPCR analysis of cartilage markers showed that undifferentiated cells expressed predominantly type IIA collagen and slightly type XI collagen (Fig. 6B). The primers used to detect type II collagen allowed to distinguish between two alternatively spliced forms, A and B. During chondrogenesis, the type II procollagen switches from type IIA to type IIB. MSCs cultured in micropellets expressed

Fig. 6. Chondrocytic differentiation in micropellet culture. Retrieved cells were spun and cultured in micropellet for 3 weeks in a BMP-2 supplemented serum-free medium. At this time, alcian blue positive staining revealed the presence of cartilaginous proteoglycans (A) in the extracellular matrix. Original magnification 50. (B) Expression of chondrocyte markers was evaluated by a RT-PCR analysis of mRNA before (D0) and after differentiation (D21). Pictures are BET-stained electrophoresis gels of PCR products. GAPDH is shown as a control for RNA sample quality.

402

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

Fig. 7. Gene transduction in mouse mesenchymal stem cells. Retrovirally transduced mouse MSCs were analysed by FACS for eGFP fluorescence. For both conditions, before and after cell labelling, 10,000 events were analysed for relative fluorescence. Dotted line: wild type cells; thick line: transduced cells. Labelled cell count and relative fluorescence value were stable with passages and after freezing thawing cycle. The data are representative of more than six different transduction experiments.

sion affected differentiation capabilities, as was demonstrated by induction of adipocytic or osteoblastic phenotypes in labelled cells (data not shown).

Discussion In this report, we present data concerning the purification of a fibroblastoid population from adult mouse bone marrow. To fully demonstrate their MSC properties, we induced cell differentiation along the three classical mesenchymal pathways, osteoblastic, adipocytic and chondrocytic lineages. Cell differentiation was characterised by matching results obtained using various techniques such as RT-PCR and cell or matrix staining. Osteoblastic differentiation was demonstrated by the expression of osteocalcin [52] and accumulation of a bone-like mineralised matrix, adipocytic differentiation by the expression of both adipsin and PPARg2 [53,54] and cytoplasmic lipid accumulation, and finally chondrocytic differentiation by the expression of both type II collagen and aggrecan [55] and accumulation of proteoglycans in the matrix. This multipotency allowed us to conclude that these cells are MSCs as defined by Pittenger et al. [9]. Our experimental approach was based on a previously published protocol [44], which described isolation of mouse stromal cells from adherent bone marrow culture, but without any further amplification of purified cells or full characterisation of differentiation potentialities. Amplification of the fibroblastoid cell population was performed by addition of cytokines EGF and PDGF, known to positively regulate CFU-f formation from whole bone marrow [56

58]. Moreover, the choice of serum may play a critical role in the purification of mouse MSCs. Indeed, we successfully amplified these cells only when we used a serum batch positively screened for optimal mouse embryonic stem cells growth. Therefore, adult MSCs may have the same fundamental needs than embryonic stem cells. Both the selected serum and growth factors chosen are probably crucial parameters for mMSCs amplification. Nevertheless, these conditions did not preclude granulo-monocytic population growth [40,43] in primary or secondary cultures, which is the main problem in mouse bone marrow MSC isolation. To overcome this problem, some authors have treated mice with 5 fluorouracile, which eliminates hematopoietic cells from the bone marrow before culture [59]. Others transduced bone marrow with vectors to give a selectable or a transformed character to MSCs [35,37,38,40]. However, in most publications describing mouse MSCs study or isolation, adherent culture of whole bone marrow was used without any further purification [41,42,60,61]. According to our results and others [40,43,62], it should be definitively considered that these unselected adherent cultures are mixed populations derived from both mesenchymal and hematopoietic stem cells. In our conditions, granulo-monocytic population progressively disappeared in about 1 month in contrast to Kitanos results that reported hematopoietic lineage persistence for at least 2 months in similar conditions [40]. This might be due to either mouse genotype (Balb/c vs. C57Bl/6J in our study; see also Ref. [43]) or growth medium as we propose that serum is highly important for mouse MSCs amplification. A vast majority of the literature reports MSCs ability to differentiate towards adipocytic, chondrocytic and osteoblastic lineages, and in some cases, towards neural lineage [63 67]. During recent years, Verfaillies team has opened up new horizons by purifying MSC-like cells from bone marrow with wider differentiation potentialities that they called multipotent adult progenitor cells (MAPCs). MAPCs are able to differentiate along the classically MSC associated pathways that we described above, but also along endothelial [5] and hepatocytic [68] pathways. These cells may therefore be very similar to embryonic stem (ES) cells. Relationships between MAPCs and MSCs are not clear. Similarly to hematopoietic progenitor hierarchy, bone marrow resident MAPC with the widest multipotency might be early progenitors of MSCs. This hypothesis is strongly supported by the acquisition of MSC membrane markers by MAPCs cultured with increased serum concentration and correlated restriction of their differentiation potentialities [20]. On the other hand, MAPCs might be issued from a rare event of de-differentiation artificially amplified in culture. In an effort to determine the identity of our mMSCs to mouse MAPCs, we analysed the expression of oct-4 and rex-1 mRNAs, two genes that are markers of embryonic stem cells and that are expressed in MAPCs. We then tried to differentiate our mMSCs along the endothelial pathway. Both results were negative

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406

403

(P.T. data not shown), suggesting that mMAPCs were not amplified in our culture conditions. The RT-PCR analysis showed that our undifferentiated cell culture expressed many genes previously described as transcriptionally active in a lineage-specific manner. Considering a classical regulatory cascade, expression of differentiation-specific markers can be greatly explained by expression of key transcriptional regulators. In the mesenchymal stem cell model C3H10T1/2, CBFA1/Runx2 can transactivate the osteopontin and type I collagen encoding genes [69]. In accordance with these results, CBFA1/ Runx2-deficient mice did not share osteoblastic differentiation and were unable to calcify their cartilaginous skeleton [70,71] illustrating the key role of this transcription factor. Therefore, CBFA1/Runx2 expression in our cells may explain part of the osteopontin and type Ia collagen gene activity. Similarly, lipoprotein lipase [72] and aP2 [73] expression could be easily explained by C/EBPa and PPARg transactivation of their respective promoters. These last transcription factors tightly regulate adipocytic genetic program [73,74]. Nevertheless, this model does not answer the major questions, how and why key transcriptional regulators of differentiation processes are still expressed in undifferentiated cell culture? In a classical hierarchical model of differentiation, a spontaneous and continuous generation of lineage restricted progenitors from stem cells present in the culture may be a first hypothesis. Alternatively, our results could be explained by the original model described by Quesenberry et al. [75] on available data concerning hematopoietic stem cell physiology, the chiaroscuro model. In this model, cells from a wide stem cell compartment oscillate between different states of differentiation competencies. Consequently, a cell may express lineage-specific genes (similarly to an unipotent progenitor) and its progeny may present a stem cell phenotype. This model underlies a great cell plasticity, which is a major MSC property. As an example, bone marrow adipocytes can dedifferentiate and redifferentiate towards adipocytic or osteoblastic pathways [76,77]. Recently, Woodbury et al. [63] demonstrated that MSCs induced to express a neuronal phenotype were able to dedifferentiate and acquire back a MSCs morphology. Moreover, they showed that clonally derived MSCs culture expressed genes normally restricted to specific lineages or embryonic germ layers. All these data and ours could take the chiaroscuro model a step further suggesting that it could also describe MSC physiology and plasticity. The growing interest in human mesenchymal stem cell purification and culture for cell therapies highlights the need for mammal models to evaluate clinical efficacy and standardise procedures. Precisely, mouse MSCs purification is a major breakthrough that will benefit from numerous genetic models of human diseases now available in this organism. In this report, we demonstrate that mouse MSCs can be purified from adult bone marrow using nonmodified mice or bone marrow. Anterior major works in the field [40,43] did

not demonstrate cell amplification or complete differentiation potentialities precluding conclusion about the stem cell nature of cells that they purified. Our cells present the same differentiation abilities as their human counterpart as they can differentiate in response to classical stimuli along adipocytic, osteoblastic and chondrocytic pathways. Moreover, we showed that they can be transduced with MMLVderived vectors for future analysis of their in vivo fate or to test gene therapy efficiency. The first evaluation of MSC transplantation using our murine MSC model in immunocompetent mice provides evidences of an immunosuppressive effect allowing tolerance of allogeneic cell grafts [78]. MSC control of immune response may allow transplantation of allogeneic MSCs or organs with neither rejection nor graft-vs.-host disease. These data illustrate well the need for a mouse model in preclinical study of MSCs therapeutic use.

Acknowledgments We thank M. Vernet for FCS providing and helpful discussion, V. Sgambato and B. Wallace for English evaluation, A. Cumano for retrovirus vector providing, P. Marie for his helpful discussion and S. Roche for his help. PT was supported by a fellowship from Association Francaise contre les Myopathies.

References
[1] E. Fuchs, J.A. Segre, Stem cells: a new lease on life, Cell 100 (2000) 143 155. [2] C.T. Jordan, I.R. Lemischka, Clonal and systemic analysis of longterm hematopoiesis in the mouse, Genes Dev. 4 (1990) 220 223. [3] M. Bhatia, J.C.Y. Wang, U. Kapp, D. Bonnet, J.E. Dick, Purification of primitive human hematopoietic cells capable of repopulating immune-deficient mice, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 5320 5325. [4] J.E. Dennis, P. Charbord, Origin and differentiation of human and murine stroma, Stem Cells 20 (2002) 205 214. [5] M. Reyes, A. Dudek, B. Jahagirdar, L. Koodie, P.H. Marker, C.M. Verfaillie, Origin of endothelial progenitors in human postnatal bone marrow, J. Clin. Invest. 109 (2002) 337 346. [6] E. Pelosi, M. Valtieri, S. Coppola, R. Botta, M. Gabbianelli, V. Lulli, G. Marziali, B. Masella, R. Muller, C. Sgadari, U. Testa, G. Bonanno, C. Peschle, Identification of the hemangioblast in postnatal life, Blood 100 (2002) 3203 3208. [7] P. Bianco, M. Riminucci, S. Gronthos, P.G. Robey, Bone marrow stromal stem cells: nature, biology and potential applications, Stem Cells 19 (2001) 180 192. [8] R.J. Deans, A.B. Moseley, Mesenchymal stem cells: biology and potential clinical uses, Exp. Hematol. 28 (2000) 875 884. [9] M.F. Pittenger, 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, Multilineage potential of adult mesenchymal stem cells, Science 284 (1999) 143 147. [10] I. Sekiya, B.L. Larson, J.R. Smith, R. Pochampally, J.-G. Cui, D.J. Prockop, Expansion of human adult stem cells from bone marrow stroma: conditions that maximize the yields of early progenitors and evaluate their quality, Stem Cells 20 (2002) 530 541.

404

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406 W. Hardy, C. Sturgeon, T. Hewett, T. Chung, W. Stock, D. Sher, S. Weissman, K. Ferrer, J. Mosca, R. Deans, A. Moseley, R. Hoffman, Mesenchymal stem cells are capable of homing to the bone marrow of non-human primates following systemic infusion, Exp. Hematol. 29 (2001) 244 255. S. Wakitani, T. Goto, S.J. Pineda, R.G. Young, J.M. Mansour, A.I. Caplan, V.M. Goldberg, Mesenchymal cell-based repair of large fullthickness defects of articular cartilage, J. Bone Joint Surg. 76 (1994) 579 592. H.A. Awad, D.L. Butler, G.P. Boivin, F.N. Smith, P. Malaviya, B. Huibregtse, A.I. Caplan, Autologous mesenchymal stem cell-mediated repair of tendon, Tissue Eng. 5 (1999) 267 277. J. Ringe, C. Kaps, B. Schmitt, K. Buscher, J. Bartel, H. Smolian, O. Schultz, G.R. Burmester, T. Haupl, M. Sittinger, Porcine mesenchymal stem cells. Induction of distinct mesenchymal cell lineages, Cell Tissue Res. 307 (2002) 321 327. J.D. Mosca, J.K. Hendricks, D. Buyaner, J. Davis-Sproul, L.C. Chuang, M.K. Majumdar, R. Chopra, F. Barry, M. Murphy, M.A. Thiede, U. Junker, R.J. Rigg, S.P. Forestell, E. Bohnlein, R. Storb, B.M. Sandmaier, Mesenchymal stem cells as vehicles for gene delivery, Clin. Orthop. 379 (2000) S71 S90 (Suppl.). H.L. Jessop, B.S. Noble, A. Cryer, The differentiation of a potential mesenchymal stem cell population within ovine bone marrow, Biochem. Soc. Trans. 22 (1994) 248S. J.E. Dennis, A. Merriam, A. Awadallah, J.U. Yoo, B. Johnstone, A.I. Caplan, A quadripotential mesenchymal progenitor cell isolated from the marrow of an adult mouse, J. Bone Miner. Res. 14 (1999) 700 709. J.-P. Remy-Martin, A. Marandin, B. Challier, G. Bernard, M. Deschaseaux, P. Herve, Y. Wei, T. Tsuji, R. Auerbach, J.E. Dennis, K.A. Moore, J.S. Greenberger, P. Charbord, Vascular smooth muscle differentiation of murine stroma: a sequential model, Exp. Hematol. 27 (1999) 1782 1795. S. Takeshita, S. Arai, A. Kudo, Identification and characterization of mouse bone marrow stromal cell lines immortalized by temperaturesensitive SV40 T antigen: supportive activity for osteoclast differentiation, Bone 29 (2001) 236 241. Y. Negishi, A. Kudo, A. Obinata, K. Kawashima, H. Hirano, N. Yanai, M. Obinata, H. Endo, Multipotency of a bone marrow stromal cell line, TBR31-2, established from ts-SV40 T antigen gene transgenic mice, Biochem. Biophys. Res. Commun. 268 (2000) 450 455. E. Arakawa, K. Hasegawa, N. Yanai, M. Obinata, Y. Matsuda, A mouse bone marrow stromal cell line, TBR-B, shows inducible expression of smooth muscle-specific genes, FEBS Lett. 481 (2000) 193 196. Y. Kitano, A. Radu, A. Shaaban, A.W. Flake, Selection, enrichment, and culture of murine mesenchymal progenitor cells by retroviral transduction of cycling adherent bone marrow cells, Exp. Hematol. 28 (2000) 1460 1469. H.K. Jin, J.E. Carter, G.W. Huntley, E.H. Schuchman, 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 (2002) 1183 1191. D.N. Kotton, B.Y. Ma, W.V. Cardoso, E.A. Sanderson, R.S. Summer, M.C. Williams, A. Fine, Bone marrow-derived cells as progenitors of lung alveolar epithelium, Development 128 (2001) 5181 5188. D.G. Phinney, G. Kopen, R.L. Isaacson, D.J. Prockop, Plastic adherent stromal cells from the bone marrow of commonly used strains of inbred mice: variations in yield, growth, and differentiation, J. Cell. Biochem. 72 (1999) 570 585. C. Kopen, D.J. Prockop, D.G. Phinney, Marrow stromal cells migrates throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 10711 10716. R.G. Hawley, F.H. Lieu, A.Z. Fong, S.J. Goldman, J.P. Leonard, T.S. Hawley, Retroviral vectors for production of interleukin-12 in the

[11] N. Querici, D. Soligo, P. Bossolasco, F. Servida, C. Lumini, G.L. Deliliers, Isolation of bone marrow mesenchymal stem cells by anti-nerve growth factor receptor antibodies, Exp. Hematol. 30 (2002) 783 791. [12] S. Gronthos, S.E. Graves, S. Ohta, P.J. Simmons, The STRO-1+ fraction of adult human bone marrow contains the osteogenic precursors, Blood 84 (1994) 4164 4173. [13] J.E. Dennis, J.-P. Carbillet, A.I. Caplan, P. Charbord, The STRO-1+ marrow cell population is multipotential, Cells Tissues Organs 170 (2002) 73 82. [14] S. Ahdjoudj, F. Lasmoles, B.O. Oyajobi, A. Lomri, P. Delannoy, P.J. Marie, Reciprocal control of osteoblast/chondroblast and osteoblast/ adipocyte differentiation of multipotential clonal human marrow stromal F/STRO-1(+) cells, J. Cell. Biochem. 81 (2001) 23 38. [15] F.P. Barry, R.E. Boyton, J.M. Murphy, S. Haynesworth, J. Zaia, The SH-3 and SH-4 antibodies recognize distinct epitopes on CD73 from human mesenchymal stem cells, Biochem. Biophys. Res. Commun. 289 (2001) 519 524. [16] F.P. Barry, R.E. Boyton, S. Haynesworth, J.M. Murphy, J. Zaia, The monoclonal antibody SH-2, raised against human mesenchymal stem cells, recognizes an epitope on endoglin (CD105), Biochem. Biophys. Res. Commun. 265 (1999) 134 139. [17] S.-C. Hung, N.-J. Chen, S.-L. Hsieh, H. Li, H.-L. Ma, W.-H. Lo, Isolation and characterization of size-sieved stem cells from human bone marrow, Stem Cells 20 (2002) 249 258. [18] A. Muraglia, R. Cancedda, R. Quarto, Clonal mesenchymal progenitors from human bone marrow differentiate in vitro according to a hierarchical model, J. Cell Sci. 113 (2000) 1161 1166. [19] D.C. Colter, I. Sekiya, D.J. Prockop, Identification of a subpopulation of rapidly self-renewing and multipotential adult stem cells in colonies of human marrow stromal cells, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 7841 7845. [20] M. Reyes, T. Lund, T. Lenvik, D. Aguiar, L. Koodie, C.M. Verfaillie, Purification and ex vivo expansion of postnatal human marrow mesodermal progenitor cells, Blood 98 (2001) 2615 2625. [21] Y. Jiang, B.N. Jahagirdar, R.L. Reinhardt, R.E. Schwartz, C.D. Keene, X.R. Ortiz-Gonzalez, M. Reyes, T. Lenvik, T. Lund, M. Blackstad, J. Du, S. Aldrich, A. Lisberg, W.C. Low, D.A. Largaespada, C.M. Verfaillie, Pluripotency of mesenchymal stem cells derived from adult marrow, Nature 418 (2002) 41 49. [22] E.M. Horwitz, D.J. Prockop, L.A. Fitzpatrick, W.W. Koo, P.L. Gordon, M. Neel, M. Sussman, P. Orchard, J.C. Marx, R.E. Pyeritz, M.K. Brenner, Transplantability and therapeutic effects of bone marrow-derived mesenchymal stem cells in children with osteogenesis imperfecta, Nat. Med. 5 (1999) 309 313. [23] E.M. Horwitz, P.L. Gordon, W.K.K. Koo, J.C. Marx, M.D. Neel, R.Y. McNall, L. Muul, T. Hofmann, Isolated allogenic bone marrow-derived mesenchymal stem cells engraft and stimulate growth in children with osteogenesis imperfecta: implications for cell therapy of bone, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 8932 8937. [24] D.J. Prockop, Marrow stromal cells as stem cells for nonhematopoietic tissues, Science 276 (1997) 71 74. [25] M.F. Pittenger, D.R. Marshak, Mesenchymal stem cells of human adult bone marrow, in: D.R. Marshak, R.L. Gardner, D. Gottlieb (Eds.), Stem Cell Biology, Cold Spring Harbor Laboratory Press, New York, 2001, pp. 349 373. [26] E.J. Schwarz, G.M. Alexander, D.J. Prockop, S.A. Azizi, Multipotential marrow stromal cells transduced to produce L-DOPA: engraftment in a rat model of Parkinson disease, Hum. Gene Ther. 10 (1999) 2539 2549. [27] D.R. Martin, N.R. Cox, T.L. Hathcock, G.P. Niemeyer, H.J. Baker, Isolation and characterization of multipotential mesenchymal stem cells from feline bone marrow, Exp. Hematol. 30 (2002) 879 886. [28] S. Kadiyala, R.G. Young, M.A. Thiede, S.P. Bruder, Culture expanded canine mesenchymal stem cells possess osteochondrogenic potential in vivo and in vitro, Cell Transplant 6 (1997) 125 134. [29] S.M. Devine, A.M. Bartholomew, N. Mahmud, M. Nelson, S. Patil,

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406 bone marrow to induce a graft-versus-leukemia effect, Ann. N. Y. Acad. Sci. 795 (1996) 341 345. L. Albright, J.C. Madigan, M.R. Gaston, D.P. Houcheus, Therapy in an intracerebral murine glioma model, using Bacillus Calmette Guerin, neuraminidase-treated tumor cells, and 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea, Cancer Res. 35 (1975) 658 665. I.K. Moutsatsos, G. Turgeman, S. Zhou, B.G. Kurkalli, G. Pelled, L. Tzur, P. Kelley, N. Stumm, S. Mi, R. Muller, Y. Zilberman, D. Gazit, Exogenously regulated stem cell-mediated gene therapy for bone regeneration, Mol. Ther. 3 (2001) 449 461. P. Tropel, V. Roullot, M. Vernet, C. Poujol, H. Pointu, P. Nurden, G. Marguerie, D. Tronik-Le Roux, A 2.7-kb portion of the 5 flanking region of the murine glycoprotein alphaIIb gene is transcriptionally active in primitive hematopoietic progenitor cells, Blood 90 (1997) 2995 3004. A.J. Friedenstein, U.J. Gorskaja, N.N. Kulagina, Fibroblast precursors in normal and irradiated mouse hematopoietic organs, Exp. Hematol. 4 (1976) 267 274. C.X. Xu, J.H. Hendry, N.G. Testa, T.D. Allen, Stromal colonies from mouse marrow: characterization of cell types, optimization of plating efficiency and its effect on radiosensitivity, J. Cell Sci. 61 (1983) 453 466. N. Jaiswal, S.E. Haynesworth, A.I. Caplan, S.P. Bruder, Osteogenic differentiation of purified, culture-expanded human mesenchymal stem cells in vitro, J. Cell. Biochem. 64 (1997) 295 312. G. Karsenty, Transcriptional control of osteoblast differentiation, Endocrinology 142 (2001) 2731 2733. K. Cianflone, M. Malowska, Differentiation-induced production of ASP in human adipocytes, Eur. J. Clin. Invest. 25 (1995) 817 825. A. Vidal-Puig, M. Jimenez-Linan, B.B. Lowell, A. Hamann, E. Hu, B. Spiegelman, J.S. Flier, D.E. Moller, Regulation of PPARg gene expression by nutrition and obesity in rodents, J. Clin. Invest. 97 (1996) 2553 2561. J.U. Yoo, T.S. Barthel, K. Nishimura, L. Solchaga, A.I. Caplan, V.M. Goldberg, B. Johnstone, The chondrogenic potential of human bonemarrow derived mesenchymal progenitors cells, J. Bone Surg. Am. 80 (1998) 1745 1757. K. Satomura, A.R. Derubeis, N.S. Fedarko, K. Ibaraki-OConnor, S.A. Kuznetsov, D.W. Rowe, M.F. Young, P. Gehron Robey, Receptor tyrosine kinase expression in human bone marrow stromal cells, J. Cell. Physiol. 177 (1998) 426 438. S.A. Kuznetsov, M.H. Mankani, S. Gronthos, K. Satomura, P. Bianco, P.G. Robey, Circulating skeletal stem cells, J. Cell Biol. 153 (2001) 1133 1139. Q.R. Wang, Z.J. Yan, N.S. Wolf, Dissecting the hematopoietic microenvironment. VI. The effects of several growth factors on the in vitro growth of murine bone marrow CFU-F, Exp. Hematol. 18 (1990) 341 347. N. Falla, P. Van Vlasselear, B. Borremans, G. Schoeters, U. Van Gorp, Characterization of a 5-fluorouracil-enriched osteoprogenitor population of the murine bone marrow, Blood 82 (1993) 3580 3591. R.F. Pereira, M.D. OHara, A.V. Laptev, K.W. Halford, M.D. Pollard, R. Class, D. Simon, K. Livezey, D.J. Prockop, Marrow stromal cells as a source of progenitor cells for nonhematopoietic tissues in transgenic mice with a phenotype of osteogenesis imperfecta, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 1142 1147. D. Gazit, Y. Zilberman, G. Turgeman, S. Zhou, A. Kahn, Recombinant TGF-h1 stimulates bone marrow osteoprogenitor cell activity and bone matrix synthesis in osteopenic, old male mice, J. Cell. Biochem. 73 (1999) 379 389. Q.R. Wang, N.S. Wolf, Dissecting the hematopoietic microenvironment. VIII. Clonal isolation and identification of cell types in murine CFU-F colonies by limiting dilution, Exp. Hematol. 18 (1990) 355 359. D. Woodbury, K. Reynolds, I.B. Black, Adult bone marrow stromal stem cells express germline, ectodermal, endodermal, and mesodermal genes prior to neurogenesis, J. Neurosci. Res. 96 (2002) 908 917.

405

[46]

[47]

[48]

[49]

[50]

[51]

[52] [53] [54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64] S.-C. Hung, H. Chen, C.-Y. Pan, M.J. Tsai, L.-S. Kao, H.-L. Ma, In vitro differentiation of size-sieved stem cells into electrically active neural cells, Stem Cells 20 (2002) 522 529. [65] J.R. Sanchez-Ramos, S. Song, F. Cardozo-Pelaez, C. Hazzi, T. Stedeford, A. Willing, T.B. Freeman, S. Saporta, W. Janssen, N. Patel, D.R. Cooper, P.R. Sanberg, Adult bone marrow stromal cells differentiate into neural cells in vitro, Exp. Neurol. 164 (2000) 247 256. [66] L.-R. Zhao, W.-M. Duan, M. Reyes, C.D. Keene, C.M. Verfaillie, W.C. Low, Human bone marrow stem cells exhibit neural phenotypes and ameliorates neurological after grafting into the ischemic brain of rats, Exp. Neurol. 174 (2002) 11 20. [67] J.R. Sanchez-Ramos, Neural cells derived from adult bone marrow and umbilical cord blood, J. Neurosci. Res. 69 (2002) 880 893. [68] R.E. Schwartz, M. Reyes, L. Koodie, Y. Jiang, M. Blackstad, T. Lund, T. Lenvik, S. Johnson, W.S. Hu, C.M. Verfaillie, Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells, J. Clin. Invest. 109 (2002) 1291 1302. [69] H. Harada, S. Tagashira, M. Fujiwara, S. Ogawa, T. Katsumata, A. Yamaguchi, T. Komori, M. Nakatsuka, Cbfa1 isoforms exert functional differences in osteoblast differentiation, J. Biol. Chem. 274 (1999) 6972 6978. [70] F. Otto, A.P. Thornell, T. Crompton, A. Denzel, K.C. Gilmour, I.R. Rosewell, G.W. Stamp, R.S. Beddington, S. Mundlos, B.R. Olsen, P.B. Selby, M.J. Owen, Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development, Cell 89 (1997) 765 771. [71] T. Komori, H. Yagi, S. Nomura, A. Yamaguchi, K. Sasaki, K. Deguchi, Y. Shimizu, R.T. Bronson, Y.H. Gao, M. Inada, M. Sato, R. Okamoto, Y. Kitamura, S. Yoshiki, T. Kishimoto, Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts, Cell 89 (1997) 755 764. [72] E. Mueller, S. Drori, A. Aiyer, J. Yie, P. Sarraf, H. Chen, S. Hauser, E.D. Rosen, K. Ge, R.G. Roeder, B.M. Spiegelman, Genetic analysis of adipogenesis through peroxisome proliferator-activated receptor g isoforms, J. Biol. Chem. 277 (2002) 41925 41930. [73] E.D. Rosen, C.J. Walkey, P. Puigserver, B.M. Spiegelman, Transcriptional regulation of adipogenesis, Genes Dev. 14 (2000) 1293 1307. [74] P. Tontonoz, E. Hu, B.M. Spiegelman, Stimulation of adipogenesis in fibroblasts by PPARg2, a lipid-activated transcription factor, Cell 79 (1994) 1147 1156. [75] P.J. Quesenberry, G.A. Colvin, J.-F. Lambert, The chiaroscuro stem cell: a unified stem cell theory, Blood 100 (2002) 4266 4271. [76] S.R. Park, R.O.C. Oreffo, J.T. Triffitt, Interconversion potential of cloned marrow adipocytes in vitro, Bone 24 (1999) 549 554. [77] J.H. Bennett, C.J. Joyner, J.T. Triffitt, M.E. Owen, Adipocytic cells culture from marrow have osteogenic potential, J. Cell Sci. 99 (1991) 131 139. [78] F. Djouad, P. Plence, C. Bony, P. Tropel, F. Apparailly, J. Sany, D. Noel, C. Jorgensen, Immunosuppressive effect of mesenchymal stem cells favors tumor growth in allogeneic animals, Blood 102 (2003) 3837 3844. [79] J.M. Taylor-Jones, R.E. McGehee, T.A. Rando, B. Lecka-Czernik, D.A. Lipschitz, C.A. Peterson, Activation of an adipogenic program in adult myoblasts with age, Mech. Ageing Dev. 123 (2002) 649 661. [80] J.M. Gimble, C.E. Robinson, X. Wu, K.A. Kelly, B.R. Rodriguez, S.A. Kliewer, J.M. Lehmann, D.C. Morris, Peroxysome proliferatoractivated receptor-g activation by thiazolidinediones induces adipogenesis in bone marrow stromal cells, Mol. Pharmacol. 50 (1996) 1087 1094. [81] J. Rentsch, M. Chiesi, Regulation of ob gene mRNA levels in cultured adipocytes, FEBS Lett. 379 (1996) 55 59. [82] S.H. Windahl, O. Vidal, G. Andersson, J.A. Gustafsson, C. Ohlsson, Increased cortical bone mineral content but unchanged trabecular bone mineral density in female ERb/ mice, J. Clin. Invest. 104 (1999) 895 901.

406

P. Tropel et al. / Experimental Cell Research 295 (2004) 395406 [84] E. Kitaoka, K. Satomura, E. Hayashi, K. Yamanouchi, S. Tobiume, K. Kume, M. Obinata, M. Nagayama, Establishment and characterization of chondrocyte cell lines from the costal cartilage of SV40 large T antigen transgenic mice, J. Cell. Biochem. 81 (2001) 571 582.

[83] I. Chackalaparampil, A. Peri, M. Nemir, M.D. Mckee, P.H. Lin, B.B. Mukherjee, A.B. Mukherjee, Cells in vivo and in vitro from osteopetrotic mice homozygous for c-src disruption show suppression of synthesis of osteopontin, a multifunctionnal extracellular matrix protein, Oncogene 12 (1996) 1457 1467.