E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m w w w. e l s e v i e r. c o m / l o c a t e / y e x c r Research Article Stem molecular signature of adipose-derived stromal cells Daniele Peroni a , Ilaria Scambi a , Annalisa Pasini b , Veronica Lisi b , Francesco Bifari b , Mauro Krampera b , Gino Rigotti c , Andrea Sbarbati a , Mirco Galiè a,⁎ a Department of Morphological and Biomedical Sciences, Section of Anatomy and Histology, University of Verona, Italy Department of Clinical and Experimental Medicine, Section of Haematology, University of Verona, Italy c Plastic and Reconstructive Surgery (II Division), City Hospital, Verona, Italy b ARTICLE INFORMATION ABS T R AC T Article Chronology: It has recently been shown that adipose tissue is an abundant and easily accessible Received 27 June 2007 source of stromal progenitor cells (ADSCs, adipose-derived stromal cells), resembling Revised version received the mesenchymal stem cells (MSCs) obtained from adult bone marrow. However, the 10 September 2007 identification of these two lineages is still controversial and even the stem cell nature Accepted 6 October 2007 of ADSCs is doubted. In this study, we examined the “stemness” transcriptional profile Available online 17 October 2007 of ADSCs and BM-MSCs, with two aims: (1) to compare their “stem cell molecular signature” and (2) to dissect their constitutive expression pattern for molecules Keywords: involved in tissue development, homeostasis and repair. As well as several Mesenchymal stem cells molecules involved in matrix remodeling and adult tissue angiogenesis and repair, Adipose tissue we detected the expression of genes UTF-1, Nodal, and Snail2, which are known to be UTF-1 expressed by embryonic stem cells but have been never described in other stem Nodal lineages. In addition, for the first time we described the transcriptional profile of Snail2 human BM-MSCs and ADSCs for the CD44 splice variants, which are determinant in CD44 splice variants cell trafficking during embryonic development, in adult tissue homeostasis and also in tumor dissemination. Thus, our findings strongly support a close relationship between ADSCs and BMMSCs, suggest an unexpected similarity between MSCs and embryonic stem cells, and possibly support the potential therapeutic application of ADSCs. © 2007 Elsevier Inc. All rights reserved. Introduction Despite a common immunophenotype [1], the identification of adipose-derived mesenchymal stem cells with bone-marrowderived mesenchymal stem cells (BM-MSCs) is still controversial, and even their stem cell nature is doubted; for this reason, the first cell type is frequently referred to by the term “adipose-derived stromal cells” (ADSCs). However, ADSCs share important biological properties with BM-MSCs, since they can be induced to myogenic [2], cardiomyogenic [3], epithelial [4], endothelial [5,6] and neurogenic [7] differentiation. In addition, both cell types may release trophic mediators such as pro-angiogenic and anti-inflammatory factors [8]. The functional analogy with BM-MSCs would seem to encourage ⁎ Corresponding author. Fax: +39 045 8027163. E-mail address: mirco@anatomy.univr.it (M. Galiè). 0014-4827/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2007.10.007 604 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 the use of ADSCs for tissue engineering and cell therapy strategies. At the same time, clarification of their nature is required through the definition of a “molecular signature” based on stem-cell-related genes, such as markers of stem lineages and genes regulating developmental and regenerative processes. A large body of literature has identified the growth factors, adhesion molecules, cytokines/chemokines and extracellular matrix proteins that drive embryonic development. Some of these molecules are known to orchestrate homeostasis and regenerative processes such as tissue turnover, revascularization of ischemic regions, and wound healing, or pathological processes such as inflammation and tumor dissemination. Several studies have addressed the molecular characterization of BM-MSCs and ADSCs [9–12], but little is known about their expression pattern for stemnessrelated molecules. The gene expression profile for these molecules is predictive of the therapeutic potential of ADSCs and provides a molecular basis for the clinical benefits recently reported in preliminary studies in murine models of human diseases [13,14]. The capacity to regulate its own motility in response to surrounding stimuli is an intrinsic aspect of stem cell phenotype, which must be understood in view of ADSC-based therapeutic applications [8]. An important adhesion molecule that plays a crucial role in cell migration/invasion programs is CD44 [15]. In addition to the standard CD44 (CD44s), many isoforms generated by the retention of alternative exons during pre-mRNA splicing have been identified. While CD44s is expressed in most mammalian cells and tissues, the splicing variants are restricted to certain cytotypes of ectodermal origen, such as keratinocytes and basal cells of squamous and glandular epithelia, or to certain physiological or pathological conditions, such as embryonic epithelial–stromal interaction [16], activation of T lymphocytes [17] and dissemination of cancer cells [18,19]. All these phenomena are based on invasive behavior and capacity to interact with the surrounding microenvironment [20]. Although MSCs are known to splice out CD44 alternative exons, the expression pattern of CD44 isoforms in human ADSCs (hADSCs) and BM-MSCs (hBM-MSCs) has never been investigated. Here we provide a high-throughput transcriptional analysis of ADSCs, with two aims: (1) to compare the ‘stem cell’ molecular signature of ADSCs with that of other stem cell populations, such as BM-MSCs, and (2) to dissect their constitutive expression pattern for molecules involved in tissue development, homeostasis and repair. Materials and methods Cell isolation hADSCs and mADSCs Human ADSCs were obtained from lipoaspirates of healthy donors after informed consent. Murine ADSCs were isolated from inguinal adipose tissues of C57BL/6 mice. For cell isolation, we used the following standard procedure: the extracellular matrix was digested at 37 °C in HBSS with 1 mg/ml collagenase type I (GIBCO life technology) and 2% BSA. After incubation, digestion enzyme activity was neutralized with Dulbecco's modified Eagle medium (DMEM) containing 20% fetal bovine serum (FBS), and centrifuged at 1200×g for 10 min to obtain a high-density pellet, which constitutes the stromal vascular fraction (SVF). It was then resuspended in 160 mM NH4Cl and incubated at room temperature for 10 min to lyse contaminating red blood cells. The SVF was collected by centrifugation and filtered through a 70-μm nylon mesh to remove cell debris. SVF was cultured in 25-cm2 flasks (BD Falcon™, Becton Dickinson, Milan, Italy) at a concentration of 1 × 105 cells/cm2, using DMEM with high glucose concentration, GLUTAMAX I™, 20% heat-inactivated fetal calf serum (FCS), 100 U/ml penicillin and 100 μg/ml streptomycin (all from GibcoBRL/ Life Technologies, Milan, Italy). hBM-MSCs Human BM-MSCs were obtained from BM aspirates of healthy donors, after informed consent. BM mononuclear cells were isolated using density gradient centrifugation (Lymphoprep, Nycomed Pharm, Oslo, Norway), as previously described [21]. Cell culture SVF and BM-MSC cultures were incubated at 37 °C in a 5% CO2 atmosphere. After 72 h, non-adherent cells were removed. When 70–80% confluent, adherent cells were trypsinized (0.05% Fig. 1 – Cytofluorimetric identification of MSCs. Human ADSCs and BM-MSCs (a) and murine ADSCs (b) were recognized on the basis of the their surface immunophenotypical profile. 605 E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 Table 1 – Stem cell related genes constitutively expressed in human ADSCs and BM-MSCs Position Gene symbol Description BM-MSC ADSC 4 5 10 23 28 30 33 35 36 37 44 45 46 47 52 53 54 62 63 ACTG2 ACVR1 BDNF BMPR2 CTNNB1 CCNE1 CD24 CD44 CD9 CDH1 CDKN1A CDKN1B CDKN2A CDKN2D COL6A2 CST3 CXCL12/SDF-1 EGF EGFR 0.39 0.63 0.42 0.59 0.40 0.40 1.00 0.63 0.30 0.54 0.30 0.49 0.29 0.49 0.39 1.02 0.51 0.28 0.37 0.35 0.62 0.26 0.43 0.30 0.29 0.92 0.46 0.33 0.48 0.31 0.49 0.30 0.48 0.40 0.92 0.40 0.29 0.42 69 74 78 84 85 105 120 121 122 126 128 134 135 141 144 148 150 FGF1 FGF16 FGF2 FGF5 FGF6 GATA4 GJB5 HSPA9B ICAM-1 IGF2 IL6 INS INSRR ITGA5 ITGA8 ITGAV ITGB1 0.28 0.29 0.54 0.34 0.32 0.34 0.30 0.56 0.23 0.30 0.45 0.68 0.37 0.85 0.23 0.30 0.51 0.34 0.30 0.54 0.33 0.35 0.28 0.23 0.38 0.27 0.35 0.50 0.62 0.37 0.54 0.30 0.31 0.50 154 166 179 180 191 192 200 210 221 224 226 227 228 229 243 245 248 249 250 257 258 260 261 262 ITGB5 MDM2 NODAL NOG PAX6 PDGFA POU5F1 PUM2 SNAI2 SOX13 SOX17 SOX18 SOX2 SOX3 THY1 TNC UTF1 VCAM-1 VEGF WNT5B WNT6 WNT7B WNT8A ZFP42 Actin, gamma 2, smooth muscle, enteric Activin A receptor, type I Brain-derived neurotrophic factor Bone morphogenetic protein receptor, type II (serine/threonine kinase) Catenin (cadherin-associated protein), beta 1, 88 kDa Cyclin E1 CD24 antigen (small cell lung carcinoma cluster 4 antigen) CD44 antigen (homing function and Indian blood group system) CD9 antigen (p24) Cadherin 1, type 1, E-cadherin (epithelial) Cyclin-dependent kinase inhibitor 1A (p21, Cip1) Cyclin-dependent kinase inhibitor 1B (p27, Kip1) Cyclin-dependent kinase inhibitor 2A (melanoma, p16, inhibits CDK4) Cyclin-dependent kinase inhibitor 2D (p19, inhibits CDK4) Collagen, type VI, alpha 2 Cystatin C (amyloid angiopathy and cerebral hemorrhage) Chemokine (C–X–C motif) ligand 12, SDF-1 (stromal cell-derived factor 1) Epidermal growth factor (beta-urogastrone) Epidermal growth factor receptor (erythroblastic leukemia viral (v-erb-b) oncogene homolog, avian) Fibroblast growth factor 1 (acidic) Fibroblast growth factor 16 Fibroblast growth factor 2 (basic) Fibroblast growth factor 5 Fibroblast growth factor 6 GATA binding protein 4 Gap junction protein, beta 5 (connexin 31.1) Heat shock 70-kDa protein 9B (mortalin-2) Intercellular adhesion molecule 1 (CD54), human rhinovirus receptor Insulin-like growth factor 2 (somatomedin A) Interleukin 6 (interferon, beta 2) Insulin Insulin receptor-related receptor Integrin, alpha 5 (fibronectin receptor, alpha polypeptide) Integrin, alpha 8 Integrin, alpha V (vitronectin receptor, alpha polypeptide, antigen CD51) Integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12) Integrin, beta 5 Mdm2, transformed 3T3 cell double minute 2, p53 binding protein (mouse) Nodal homolog (mouse) Noggin Paired box gene 6 (aniridia, keratitis) Platelet-derived growth factor alpha polypeptide POU domain, class 5, transcription factor 1 Pumilio homolog 2 (Drosophila) Snail homolog 2 (Drosophila) SRY (sex determining region Y)-box 13 SRY (sex determining region Y)-box 17 SRY (sex determining region Y)-box 18 SRY (sex determining region Y)-box 2 SRY (sex determining region Y)-box 3 Thy-1 cell surface antigen Tenascin C (hexabrachion) Undifferentiated embryonic cell transcription factor 1 Vascular cell adhesion molecule 1 Vascular endothelial growth factor Wingless-type MMTV integration site family, member 5B Wingless-type MMTV integration site family, member 6 Wingless-type MMTV integration site family, member 7B Wingless-type MMTV integration site family, member 8A Zinc finger protein 42 0.42 0.30 0.39 0.39 0.28 0.27 0.57 0.31 0.31 0.31 0.31 1.00 0.34 0.38 0.40 0.33 0.66 0.26 0.32 0.30 0.79 0.31 0.86 0.35 0.31 0.41 0.38 0.42 0.32 0.32 0.38 0.33 0.28 0.34 0.25 0.92 0.30 0.38 0.37 0.31 0.52 0.28 0.29 0.26 0.69 0.27 0.76 0.29 606 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 Fig. 2 – Stem-related transcriptional profile of hADSCs and hBM-MSCs. (a) Gene arrays are represented as the average of the arrays of long-term and short-term cultures of hADSCs and BM-MSCs, respectively. (b) The scatterplot shows that none of the genes was overexpressed more than 2-fold (green lines) in one lineage as compared to the other. trypsin at 37 °C for 5 min, GibcoBRL/Life Technologies), harvested and washed with medium to remove trypsin, and expanded in larger flasks. A homogenous cell population of ADSCs or BM-MSCs was obtained after 2 to 3 weeks of culture. Cells at early (2 to 5, short-term cultures) and late (5 to 20, longterm cultures) passage in culture were used for the experiments. HADSCs and hBM-MSCs were identified on the basis of their expression of CD105 (endoglin), CD73, CD29, CD44 and CD90, and the lack of hematopoietic (CD45, CD14, CD34) and endothelial cell markers (CD31), all assessed using cytofluorimetric analysis, as previously described [21] (Fig. 1a). The identification of mADSCs was based on the expression of CD106 and the lack of hematopoietic (CD45, CD14, CD34) and endothelial cell (CD31) marker expression, as previously described [22] (Fig. 1b). Both ADSCs and BM-MSCs were cultured in DMEM containing 20% fetal bovine serum(FBS). Cultures were carried on in parallel by using the same FBS lot. Fig. 3 – Confirmation of expression of the selected genes by RT-PCR. mRNA from BM-MSCs and ADSCs subjected to retrotranscription as described in Materials and methods. PCR amplification with specific primers confirmed the expression of genes implicated in matrix remodeling, cell migration (Integrin αV, Integrin β5, Integrin β1, CD44s) and in embryogenesis (Pou5f1, UTF-1, Nodal, Snail2). E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 CFU-F assay Three aliquots of 1 × 105 SVF cells for each of 7 patients were seeded in three 10-cm-diameter cell culture dishes, and after 10 days of culture the number of adherent cells was estimated by colony forming unit fibroblast (CFU-F) count. The culture layer was fixed in 4% formalin for 15 min and than stained by GIEMSA. Only the colonies macroscopically visible as blue spot were counted. 607 Stemness-related gene array was performed on short-term and long-term cultures of one lineage of hADSCs and one lineage of BM-MSCs. HADSCs and hBM-MSCs were obtained from different patients. RT-PCRs were performed using RNA from different lineages with respect to that used for arrays. Murine arrays were performed in duplicate using shortand long-term cultures of two distinct lineages of mADSCs, each obtained from a different animal. Experimental procedure Microarray analysis Samples Since several authors have reported some transcriptional changes upon in vitro culture of MSCs [23,24], transcriptional analysis was performed on short-term and long-term cultured human ADSCs and BM-MSCs. The expression values were reported as the average of the values obtained from different time-cultured samples of the same lineage (Table 1). The different types of mRNA extracted from each cell population were used as templates to generate a cDNA library. cRNA labeled with UTP–biotin (Enzo Roche Molecular Biochemicals, Mannheim, Germany) was retrotranscripted and purified. UTP–biotin–cRNA from human MSCs was hybridized on Gene Arrays containing probes specific for human genes related to stem phenotype, adhesion molecules and cytokines/chemokines (from SuperArray® Bioscience Corporation, Frederick, MD, USA; Cat. nos. HS-601.2). The UTP-biotin-cRNA from murine MSCs was hybridized in Gene Fig. 4 – Microarray-based transcriptional analysis of C57BL/6 ADSCs for adhesion molecules and inflammatory cytokines. (a) Gene arrays are represented as the average of the arrays of long-term and short-term cultured cells, respectively. 608 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 Table 2 – Adhesion molecules genes constitutively expressed in murine ADSCs Position 1 3 6 8 11 15 16 19 22 23 24 25 28 29 30 31 32 34 35 36 37 38 39 40 43 46 47 48 50 55 56 57 69 75 77 83 85 87 94 98 99 101 103 104 105 108 109 110 112 113 Gene symbol Gene description Adamts2 A disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motif, 2 Ctnna1 Catenin (cadherin associated protein), alpha 1 Ctnnb1 Catenin (cadherin associated protein), beta 1, 88 kDa Cdh2 Cadherin 2 Cntn1 Contactin 1 Col11a1 Procollagen, type XI, alpha 1 Col1a1 Procollagen, type I, alpha 1 Col2a1 Procollagen, type II, alpha 1 Col3a1 Procollagen, type III, alpha 1 Col4a1 Procollagen, type IV, alpha 1 Col4a2 Procollagen, type IV, alpha 2 Col5a1 Procollagen, type V, alpha 1 Col5a3 Procollagen, type V, alpha 3 Col6a1 Procollagen, type VI, alpha 1 Col6a2 Procollagen, type VI, alpha 2 Col8a1 Procollagen, type VIII, alpha 1 Cspg2 Chondroitin sulfate proteoglycan 2 Ctgf Connective tissue growth factor Ecm1 Extracellular matrix protein 1 Emilin1 Elastin microfibril interfacer 1 Entpd1 Ectonucleoside triphosphate diphosphohydrolase 1 Fbln1 Fibulin 1 Fn1 Fibronectin 1 Icam1 H1 histone family, member 0 Itga3 Integrin alpha 3 Itga4 Integrin alpha 4 Itga5 Integrin alpha 5 (fibronectin receptor alpha) Itga7 Integrin alpha 7 Itgav Integrin alpha V Itgax Integrin alpha X Itgb1 Integrin beta 1 (fibronectin receptor beta) Lamb1-1 Laminin B1 subunit 1 Mmp12 Matrix metalloproteinase 12 Mmp14 Matrix metalloproteinase 14 (membrane-inserted) Mmp2 Matrix metalloproteinase 2 Mmp23 Matrix metalloproteinase 23 Mmp3 Matrix metalloproteinase 3 Postn Periostin, osteoblast specific factor Sgce Sarcoglycan, epsilon Sparc Secreted acidic cysteine rich glycoprotein Spp1 Secreted phosphoprotein 1 Tgfbi Transforming growth factor, beta induced Thbs1 Thrombospondin 1 Thbs2 Thrombospondin 2 Timp1 Tissue inhibitor of metalloproteinase 1 Timp2 Tissue inhibitor of metalloproteinase 2 Timp3 Tissue inhibitor of metalloproteinase 3 Tnc Tenascin C Vcam1 Vascular cell adhesion molecule 1 0.58 0.60 0.74 0.31 0.31 0.35 0.88 0.34 0.79 0.90 0.35 1.02 0.34 0.77 0.41 0.50 0.38 0.90 0.71 0.51 0.32 0.43 1.16 0.32 0.33 0.33 0.39 0.37 0.32 0.48 0.44 0.84 0.30 0.82 0.40 0.32 0.75 0.77 0.31 1.15 1.03 0.34 0.35 0.55 0.63 0.71 0.51 0.32 0.53 Arrays containing probes for adhesion molecules (from SuperArray® Bioscience Corporation, Frederick, MD, USA; Cat. no. OMM-013) or cytokines/chemokines (from SuperArray® Bioscience Corporation, Frederick, MD, USA; Cat. no. OMM-011) specific genes. The hybridization pattern was revealed by CDP-Star® substrate fluorescence using Chemiluminescent Detection Kit and recorded on X-ray film. All the steps of the procedure were performed using reagent kits purchased from SuperArray® Bioscience Corporation (Frederick, MD, USA) and closely following the manufacturer's instructions. The images were acquired as a 300 dpi TIFF file using a desktop scanner and saved as a grayscale TIFF file for densitometric analysis. Analysis Data from arrays were analyzed using GEArray Expression Analysis Suite (Superarray Bioscience Corporation). Quantitative expression values were corrected to the background, normalized with respect to the positive control genes included in the arrays, and reported as ratios to the mean values of normalization genes. The maximal value of the blank spots for each array (0.29 for human stemness array; 0.30 for murine adhesion molecule array; 0.42 for murine cytokine/chemokine array) was used as a threshold for positive expression. For human stemness array, genes with values b0.29 were considered “not expressed”; genes with values ≥0.29 but ≤0.60 were considered “moderately expressed”; genes with values =0.60 were considered “significantly expressed.” For comparative analysis of the gene expression profiles of hBM-MSCs and hADSCs, we selected only genes which were “moderately and significantly expressed” (i.e., with values of expression ≥0.30) at least by one of the two lineages. Genes were considered significantly over-expressed in one lineage when their expression values proved N1.5-fold Table 3 – Cytokines/Chemokines genes constitutively expressed in murine ADSCs Position Gene symbol 2 3 9 19 33 36 40 68 80 81 82 91 Blr1 C3 Ccl2 Ccl7 Cxcl1 Cxcl12 Cxcl2 Il1r1 Il6 Il6ra Il6st Mif 96 108 Spp1 Tnfrsf1a 111 121 122 123 124 Tollip Rps27a B2m Hspcb Hspcb Burkitt lymphoma receptor 1 Complement component 3 Chemokine (C–C motif) ligand 2 Chemokine (C–C motif) ligand 7 Chemokine (C–X–C motif) ligand 1 Chemokine (C–X–C motif) ligand 12 Chemokine (C–X–C motif) ligand 2 Interleukin 1 receptor, type I Interleukin 6 Interleukin 6 receptor, alpha Interleukin 6 signal transducer Macrophage migration inhibitory factor Secreted phosphoprotein 1 Tumor necrosis factor receptor superfamily, member 1a Toll interacting protein Ribosomal protein S27a Beta-2 microglobulin Heat shock protein 1, beta Heat shock protein 1, beta 0.42 1.04 1.12 0.50 0.68 0.43 0.54 0.51 0.45 0.43 0.72 0.81 0.89 0.42 0.46 1.00 1.00 0.80 0.82 609 E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 Table 4 – Culture-associated transcriptional changes for stemness-related genes in human cells Symbol Description Ratio Genes over-expressed in short-term vs. long-term human ADSCs CDKN1B Cyclin-dependent kinase inhibitor 1B 1.88 (p27, Kip1) INS Insulin 2.18 1.63 ITGA5 Integrin, alpha 5 (fibronectin receptor, alpha polypeptide) NOG Noggin 2.00 UTF1 Undifferentiated embryonic cell 1.66 transcription factor 1 1.77 WNT6 Wingless-type MMTV integration site family, member 6 1.93 WNT8A Wingless-type MMTV integration site family, member 8A Genes under-expressed in short-term vs. long-term human ADSCs ACTG2 ATP-binding cassette, sub-family G (WHITE), member 2 ACVR1 Activin A receptor, type I BMPR2 Bone morphogenetic protein receptor, type II (serine/threonine kinase) CTNNB1 Catenin (cadherin-associated protein), beta 1, 88 kDa CCNE1 Cyclin E1 CD44 CD44 molecule (Indian blood group) CD9 CD9 molecule CDH1 Cadherin 1, type 1, E-cadherin (epithelial) COL6A2 Collagen, type VI, alpha 2 HSPA9 Heat shock 70-kDa protein 9 (mortalin) IL6 Interleukin 6 (interferon, beta 2) ITGA8 Integrin, alpha 8 ITGB1 Integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12) ITGB5 Integrin, beta 5 MDM2 Mdm2, transformed 3T3 cell double minute 2, p53 binding protein (mouse) PTEN Phosphatase and tensin homolog (mutated in multiple advanced cancers 1) PUM2 Pumilio homolog 2 (Drosophila) SNAI2 Snail homolog 2 (Drosophila) TGFBR1 Transforming growth factor, beta receptor I (activin A receptor type II-like kinase, 53 kDa) VEGFA Vascular endothelial growth factor A 0.38 0.18 0.29 0.39 0.62 0.30 0.66 0.63 0.65 0.28 0.23 0.39 0.23 0.44 0.20 0.43 0.33 0.27 0.26 Table 4 (continued) Symbol Description Ratio Genes under-expressed in short-term vs. long-term human BM-MSCs ACTG2 Actin, gamma 2, smooth muscle, enteric 0.31 ACVR1 Activin A receptor, type I 0.17 BDNF Brain-derived neurotrophic factor 0.32 BMPR2 Bone morphogenetic protein receptor, type II 0.18 (serine/threonine kinase) 0.27 CTNNB1 Catenin (cadherin-associated protein), beta 1, 88 kDa 0.52 CTNND2 Catenin (cadherin-associated protein), delta 2 (neural plakophilin-related arm-repeat protein) CCNE1 Cyclin E1 0.31 CCNE2 Cyclin E2 0.44 CD44 CD44 molecule (Indian blood group) 0.22 CDH1 CD44 molecule (Indian blood group) 0.40 0.58 CDKN1A Cyclin-dependent kinase inhibitor 1A (p21, Cip1) 0.55 CDKN2A Cyclin-dependent kinase inhibitor 2A (melanoma, p16, inhibits CDK4) COL6A2 Collagen, type VI, alpha 2 0.47 CXCL12 Chemokine (C–X–C motif) ligand 12 0.38 (stromal cell-derived factor 1) FGF2 Fibroblast growth factor 2 (basic) 0.43 HSPA9 Heat shock 70-kDa protein 9 (mortalin) 0.25 IL6 Interleukin 6 (interferon, beta 2) 0.22 0.45 ITGAV Integrin, alpha V (vitronectin receptor, alpha polypeptide, antigen CD51) ITGB1 Integrin, beta 1 (fibronectin receptor, beta 0.29 polypeptide, antigen CD29 includes MDF2, MSK12) ITGB5 Integrin, beta 5 0.41 MDM2 Mdm2, transformed 3T3 cell double 0.40 minute 2, p53 binding protein (mouse) NCAM2 Neural cell adhesion molecule 2 0.50 PUM2 Pumilio homolog 2 (Drosophila) 0.63 SNAI2 Snail homolog 2 (Drosophila) 0.30 higher than the expression of the same gene in the other lineage. On the contrary, genes were considered significantly under-expressed in one lineage when their expression values were b0.75-fold that of the expression of the same gene in the other lineage. RT-PCR 0.54 Genes over-expressed in short-term vs. long-term human BM-MSCs FGF16 Fibroblast growth factor 16 1.52 FGF6 Fibroblast growth factor 6 1.79 INS Insulin 2.07 INSRR Insulin receptor-related receptor 1.93 NODAL Nodal homolog (mouse) 1.81 NOG Noggin 2.59 SOX3 SRY (sex determining region Y)-box 3 1.52 2.05 UTF1 Undifferentiated embryonic cell transcription factor 1 WNT6 Wingless-type MMTV integration site family, 1.56 member 6 1.67 WNT7A Wingless-type MMTV integration site family, member 7A The expression of genes of interest was confirmed by RT-PCR. One hundred nanograms of mRNA from each cell population underwent reverse transcription (RT) in a reaction volume of 20 μl and in the presence of random primers (Invitrogen). One microliter of cDNA was then amplified by PCR, using the ReactionReady HotStart PCR Mix and the appropriate primer sets, both purchased from Superarray. Gene-specific primers used in this study were as follows: Nodal (PPH01944A9), UTF-1 (PPH02391A), Snail2 (Slug) (PPH02475A), Integrin β5 (PPH00634A), Integrin αV (PPH00628A), and Integrin β1 (PPH00650A). The PCR conditions were as follows: an initial denaturation of 15 min at 94 °C, 35 cycles (or 40 for Nodal and UTF-1) at 94 °C for 15 s, 59 °C for 30 s, 72 °C for 30 s, and a final extension of 5 min at 72 °C. Pou5f1 was amplified by using the primers: Fw—5′-GAC AAC AAT GAG AAC CTT CAG GAG A-3′; Rw—5′-CTG GCG CCG GTT ACA GAA CCA-3′. The mixture was first heated at 94 °C 610 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 for 2 min. Amplification was performed for 35 cycles at 94 °C for 45 s, 55 °C for 30 s and 72 °C for 90 s, followed by 72 °C for 5 min. PCR products were further analyzed on 2% agarose gels and photographed with a CCD camera. 611 expected in human patients, this quantitative estimation demonstrated that SFV of human lipoaspirates contains an amount of CFU-F which is 400 times greater than that of bone marrow, where CFU frequency has been estimated at around 1 per 105 cells (corresponding to a few hundred MSCs per cubic centimeter of marrow) [26]. Analysis of CD44 splice variants by exon-specific PCR To assess the expression of different isoforms of CD44 in MSCs, the mRNA was retrotranscripted as described above. One microliter of resulting cDNA was subjected to PCR using Taq buffer 1× (Promega), 1.5 mM MgCl2, 0.2 mM dNTPs, 1 U of Taq polymerase (Promega) and 0.4 μM of each specific primer. The standard CD44 (CD44s) was detected using primers corresponding to 5′ (hSF 5′-AAGACATCTACCCCAGCAAC-3′) and 3′ (hSR 5′-CCAAGATGATCAGCCATTCTGG-3′) constant domains. The variant isoforms were amplified using a set of 5′ primers that specifically match to variable exons, and the common 3′ primer hSR. The exon-specific primer sequences were the same as used by Van Weering et al. [25]: pv2 5′-GATGAGCACTAGTGCTACAG-3′; pv3I 5′-ACGTCTTCAAATACCATCTC-3′; pv3II 5′-TGGGAGCCAAATGAAGAAAA-3′; pv4 5′-TCAACCACACCACGGGCTTT-3′; pv5 5′-GTAGACAGAAATGGCACCAC-3′; pv6 5′-GAGGCAACTCCTAGTAGTAC-3′; pv7 5′-CAGCCTCAGCTCATACCAGC-3′; pv8 5′-TCCAGTCATAGTACAACGCT3′; pv9 5′-CAGAGCTTCTCTACATCACA-3′; pv10 5′-GGTGGAAGAAGAGACCCAAA-3′. We omitted the exon v1 analysis because it is not expressed in human cells. Furthermore, we used two different primer pairs for exon v3 because it displays two alternative splicing forms (CD44v3I and CD44v3II). PCR conditions were as follows: 94 °C for 2 min, 55 °C for 2 min, 72 °C for 2 min (1×); 94 °C for 45 s, 55 °C for 1 min, 72 °C for 1 min (35×); 72 °C for 5 min. Amplification products were resolved on 2% agarose gels. Statistical analysis To statistically evaluate the similarity of molecular signatures in ADSCs and BM-MSCs we calculated the correlation index. P b 0.05 was taken to be statistically significant. Human ADSCs and BM-MSCs have analogous transcriptional phenotypes Transcriptional analysis of hBM-MSCs and hADSCs revealed a virtually identical profile of genes identifying the stem cell phenotype (Fig. 2a). The list of genes included in the analysis and the quantification of their transcriptional expression are detailed in the Supplementary Data. The correlation coefficient between hADSCs and hBMMSCs was 0.93 (P b 0.01), indicating that the gene expression profiles were very similar. None of the genes was overexpressed more than 2-fold in one lineage compared to the other (Fig. 2b). Taking into account only the “expressed” genes (see Materials and methods), the only gene that displayed a significantly different expression (N1.5-fold) in the two kinds of MSCs was brain-derived neurotrophic factor (BDNF). It was moderately expressed in hBM-MSCs (0.42) but was not detected in hADSCs. The consistency of the results obtained using different sources and donors suggests that the transcriptional phenotype described here was not due to individual variability and can be generalized. The ADSC molecular signature includes genes typical of other stem lineages Interestingly, both hBM-MSCs and hADSCs from different patients and at different culture time points expressed molecules that are typical of embryonic phenotype, such as Oct4, and that have never been described in other stem lineages so far, i.e., undifferentiation transcription factor (UTF-1) and Nodal (Figs. 2a and 3; Table 1). These molecules are crucial for the migration of embryonic precursors during development; in addition, they are necessary to maintain the undifferentiated status of embryonic stem cells in culture. Results Regenerative profile of hBM-MSCs and hADSCs Quantitative estimate of CFU-F population in human lipoaspirates On the basis of the CFU-F assays, we estimated that 1 ml of human lipoaspirate contained at least 7.3 × 103 ± 6.3 × 103 Cells capable of forming fibroblast-like colonies, corresponding to 0.5–5% of the total SVF population (4.04 ± 3.26 on average). Despite the high individual variability observed, which is to be The genes that were constitutively expressed in both hBMMSCs and hADSCs are listed in Table 1. This list included genes that are involved in tissue development, homeostasis and repair, such as adhesion molecules (integrin α5, integrin αV, Integrin β1, Integrin β5), cytokines/chemokines (Cxcl 12/SDF-1, Interleukin 6), cyclin-dependent kinase inhibitors (CDK 1A/ p21Cip1, CDK 1B/p27Kip1, CDK 2A/p16, CDK 2D/p19), growth Fig. 5 – Transcriptional changes with culture. In order to assess culture-associated transcriptional changes we compare transcriptome of short- and long-term cultured cells. Hierarchical clustergrams for stemness-related molecules in human cells (a), adhesion molecules (b) and cytokines/chemokines (c) in murine cells demonstrated an higher statistical correlation between cells at the same time points of culture than between cells at different time points of culture. Semi-quantitative RT-PCR (d) confirmed the increased expression of CD44 in long-term (LT) than in short-term (ST) cultured human cells. UTF-1 and Nodal resulted undetectable by RT- PCR in short-term cultured cells. 612 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 factors (FGF-2, FGF-16, FGF-5, FGF-6, VEGF, PDGF-A), bone morphogenetic protein receptors (BMPR-2) and Winglesstype MMTV integration site (WNT-6, WNT8A) family members. Moreover, we observed the expression of Snail2 (Slug), a key molecule in the migration of neural-crest-derived cells inside the embryonic mesoderm and which is associated with the epithelial–mesenchymal transition in cancer metastasis (Figs. 2a and 3; Table 1). The expression of genes of interest was confirmed by RTPCR (Fig. 3). Adhesion molecules and cytokine/chemokine profile of murine ADSCs (mADSCs) Assessment of the therapeutic potential of ADSCs requires pre-clinical studies with animal models. For this reason, we extended our study to the transcriptional analysis of murine ADSCs. In particular, we investigated the expression pattern of adhesion molecules and cytokines/chemokines of C57BL/6 ADSCs (Fig. 4; Tables 2 and 3). In comparison to hBM-MSCs and hADSCs, the transcriptional analysis of adhesion molecules and cytokine/chemokines in mADSCs confirmed the expression of integrin α5, αV and β1, catenin β1, and IL-6. Moreover, we found some expressed molecules which are involved in cell migration inside the tissues, such as matrix metalloproteinases (MMP-12, MMP-14, MMP-2, MMP-23, MMP-3), tissue inhibitors of MMPs (TIMP-1, TIMP-2, TIMP-3), several isoforms of procollagen, and a number of chemokines/cytokines that play a role in chemotaxis (Ccl2/ MCP-1, Ccl7/MCP-3, Cxcl1/Gro-α, Cxcl2/Gro-β). Transcriptional changes with culture In order to evaluate transcriptional changes with in vitro culture, we focused on genes that were proven to be more than 1.5-fold over-expressed in short-term than in long-term cultured cells or vice versa. Hierarchical correlation clustergrams demonstrated the consistency of the results obtained, as they have shown high correlation between short-term as well as long-term cultures, despite their origen from different tissue (human cells) and animals (murine cells) (Fig. 5A). Accordingly with data previously reported in literature [23,27] we found an increased expression of mesenchymal stem cell-associated markers (CD44, Integrin β1/CD29) in longterm than in short-term cultured human and murine ADSCs (Tables 4 and 5). An inverse result was obtained for the markers of undifferentiated status such as UTF-1 and Nodal, which resulted over-expressed in short-term than in long-term cultured cells, indicating a shift toward a differentiated status with the in vitro culture. Semi-quantitative RT-PCR confirmed these results, as CD44 resulted over-expressed in long-term cultured cells, while UTF-1 and Nodal resulted detectable only in short-term but not in long-term ADSCs and BM-MSCs (Fig. 5b). Interestingly, in murine ADSCs we found that a prolonged culture resulted in a significant downregulation of varied isoforms of Procollagen, matrix metallopeptidases and inflammatory cytokines (Table 5). Table 5 – Culture-associated transcriptional changes for adhesion molecules and cytokines/chemokines genes in murine cells Symbol Description Ratio Adhesion molecule genes over-expressed in short-term vs. long-term murine ADSCs Adamts2 A disintegrin-like and metalloprotease 3.24 (reprolysin type) with thrombospondin type 1 motif, 2 Ctnna2 Catenin alpha 2 1.87 Cdh5 Cadherin 5 1.68 Cntn1 Contactin 1 1.62 Col11a1 Procollagen, type XI, alpha 1 1.54 Col2a1 Procollagen, type II, alpha 1 1.71 Col3a1 Procollagen, type III, alpha 1 2.74 Col5a3 Procollagen, type V, alpha 3 1.89 Col6a1 Procollagen, type VI, alpha 1 2.75 Col6a2 Procollagen, type VI, alpha 2 1.89 Fbln1 Fibulin 1 1.54 Mmp12 Matrix metalloproteinase 12 1.68 Mmp2 Matrix metalloproteinase 2 1.63 Mmp3 Matrix metalloproteinase 3 3.66 Adhesion molecule genes under-expressed in short-term vs. long-term murine ADSCs Cd44 CD44 antigen 0.56 Cdh2 Cadherin 2 0.64 Ctgf Cadherin 2 0.63 Itgb1 Integrin beta 1 (fibronectin receptor beta) 0.51 Timp3 Tissue inhibitor of metalloproteinase 3 0.40 Tnc Tenascin C 0.62 Cytokines and chemokines genes over-expressed in short-term vs. long-term murine ADSCs Blr1 Burkitt lymphoma receptor 1 C3 Complement component 3 Ccl2 Chemokine (C–C motif) ligand 2 Ccl8 Chemokine (C–C motif) ligand 8 Cxcl1 Chemokine (C–X–C motif) ligand 1 Cxcl10 Chemokine (C–X–C motif) ligand 10 Cxcl2 Chemokine (C–X–C motif) ligand 2 Cxcl4 Chemokine (C–X–C motif) ligand 4 Ifng Interferon gamma Il12rb2 Interleukin 12 receptor, beta 2 1.76 2.98 1.67 1.55 1.93 1.50 2.71 1.66 1.56 1.68 Cytokines and chemokines genes under-expressed in short-term vs. long-term murine ADSCs None MSCs show a similar expression pattern of CD44 splice variants CD44 receptor is a typical marker of MSCs. In addition to the standard isoform (CD44s), CD44 can exist in many alternative splice variants, whose expression is restricted to a small number of cytotypes (Fig. 6a). We performed exon-specific PCR of the CD44 isoforms with human ADSCs and BM-MSCs obtained from different patients and cultured for different periods (Fig. 6b). As shown in Fig 6c, we found some differences in band intensity between different samples. However, most samples showed a similar splicing pattern, including the variants v2 (280 bp), v3I (280 bp), v3II (260 bp), v5 (270 bp), v6 (280 bp), v7 (290 bp), v8 (260 bp), v9 (240 bp), v10 (340 bp) and the combinations v4–5 (380 bp), v7–10 (680 bp), v8– E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 613 Fig. 6 – Expression profile of CD44 isoforms. (a) The CD44 pre-mRNA contains at least 20 exons, ten of which are constitutively expressed and encode the standard isoform (CD44s). The remaining exons (v1–v10) can be alternatively spliced to generate over 700 different isoforms (CD44v). Exon-specific RT-PCR (b) demonstrated that ADSCs and BM-MSCs show a similar CD44 expression pattern (c). 9 (340 bp), v8–10 (540 bp) and 9–10 (430 bp). Furthermore, after short-term culture ADSCs have additional splicing variants in comparison with long-term cultures, including v5–10 (930 bp), and v6–10 (810 bp). Discussion In this study we showed that BM-MSCs and ADSCs exhibit a virtually identical transcriptional profile for genes related to the stem cell phenotype, supporting the characterization of ADSCs as a peripheral lineage of MSCs. Interestingly, we found that BM-MSCs and ADSCs express various genes shared by embryonic stem cells. Along with Pou5f1 (Oct4), which has been described to be expressed in MSCs [28], we found UTF-1, a common marker of embryonic stem cells, and Nodal which represents key signal in the embryonic stem cell system. UTF-1 and Nodal mRNAs are normally found in embryonic stem lineages and in germ line tissues, but their expression is rapidly downregulated after cell fate determination, so that their disappearance represents an early sign of the differentiation process [29]. Our study is the first to show the expression of UTF-1 and Nodal transcripts, in adult stem cells also, thus supporting a close relationship between MSCs and embryonic stem cells. The molecular analogy with BM-MSCs and even embryonic stem cells supports the use of ADSCs for diverse therapeutic applications, such as plastic surgery, tissue repair and revascularization of ischemic tissues, and provides a molecular basis for the clinical benefits recently reported in preliminary studies on murine models of human pathologies [13]. A large body of literature concerns the mitogenic, angiogenic and differentiative signals involved in these mechanisms. We have shown here that different molecules involved in these signaling pathways are constitutively expressed in ADSCs as well as in BM-MSCs. Besides basic FGF (FGF2), VEGF and PDGF-A, whose involvement in the recruitment of endothelial cell precursors is well established, we found the Cxcl12/SDF-1 gene expressed. The chemokine Cxcl12/SDF-1 has recently been shown to regulate the trafficking of hematopoietic stem cells (HSCs) from the bone marrow to peripheral blood and to enhance post-ischemia angiogenesis [30]. In addition, Cxcl12/SDF-1 is involved in embryonic development, as it is a crucial regulator of the early and late phases of embryogenesis [31,32] and it plays a role in the migration of germ cell precursors [33]. In addition to multilineage plasticity and pro-angiogenic potential, the capacity to home in on their putative niche through tissue barriers is an intrinsic aspect of stem cell phenotype. Therefore, identification of molecules involved in cell mobilization during embryonic development, inflammation, wound healing and tumor dissemination is useful in predicting the therapeutic potential of ADSCs [8,14] and the possible risks associated with their use. In our study of mADSCs, we observed the expression of several matrix metalloproteinases (MMP-12, MMP-14, MMP-2, MMP-23, MMP-3) and tissue inhibitors of MMPs (TIMP-1, TIMP-2, TIMP3) which contribute to physiological remodeling of the extracellular matrix. In particular, MMP-2 and TIMP-2 are essential for the invasive capacity of BM-MSCs [34]. Among the invasion/migration molecules detected, Snail2 (Slug) is also of interest. Snail2 is a zinc finger transcription factor which plays a prominent role in embryonic development through the regulation of neural crest migration [35]. It also 614 E XP ER I ME NT A L C EL L RE S EA R CH 3 14 ( 20 0 8 ) 6 0 3 –61 5 has an important role in pathological conditions since it can confer motile phenotype to neoplastic cells [36,37]. Its constitutive expression in MSCs suggests that Snail2 is a key regulator of the physiological properties of this stem cell populations, as well as of the predisposition of stem cells to spontaneous transformation [38]. Like Snail2, CD44 splice variants (CD44v) are widely studied for their role in cell growth and trafficking in both physiological and pathological conditions. Although CD44 represents one of the main identifying markers of the MSCs, and its role in MSCs migration has been well-documented [39], the expression pattern of CD44 splice variants in ADSCs has never been investigated so far. Our results demonstrate that CD44 is alternatively spliced in BM-MSCs and ADSCs, leading to similar expression patterns in the two cell types, which do not vary significantly during in vitro expansion. Thus, the heterogeneous expression of CD44 isoforms may contribute to the plasticity of MSCs and their interaction with the microenvironment, which are both typical aspects of stem cell phenotype. Taken together, our results support the strong similarity of ADSCs and BM-MSCs, suggesting that ADSCs may be as promising as BM-MSCs for various therapeutic applications, such as hematopoietic stem cell transplantation [40,41], tissue regeneration [42], treatment of severe acute graft versus host disease [43] and autoimmune disorders [44]. However, we found that prolonged in vitro culture significantly alters the transcriptional phenotype of both ADSCs and BM-MSCs. In accordance with data reported by other authors we shown an increased expression of mesenchymal stem-associated adhesion molecules, such as CD44 and integrinB1/CD29 both in human than in murine cells. Moreover, we found the drastic downregulation of extracellular matrixassociated molecules (varied procollagen isoforms, MMPs and TIMPs). The culture-associated changes in adhesion molecule profile reasonably reflect the adaptation to the new microenvironment. It should be a relevant insight to assess if these changes are reversible or if they stably affects the stem cell physiology. Of notice, we found also the downregulation of embryonic markers, such as UTF1 and Nodal, with culture. Acknowledgments We thank Dr. Paolo Farace for the graphic elaboration of gene array images. This work was supported by Cassa di Risparmio di Verona, Vicenza, Belluno e Ancona. [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.yexcr.2007.10.007. [15] [16] REFERENCES [17] [1] P.A. Zuk, M. Zhu, P. Ashjian, D.A. De Ugarte, J.I. Huang, H. Mizuno, Z.C. Alfonso, J.K. Fraser, P. Benhaim, M.H. Hedrick, Human adipose tissue is a source of multipotent stem cells, Mol. Biol. Cell 13 (2002) 4279–4295. H. Mizuno, P.A. Zuk, M. Zhu, H.P. Lorenz, P. Benhaim, M.H. Hedrick, Myogenic differentiation by human processed lipoaspirate cells, Plast. Reconstr. Surg. 109 (2002) 199–209. K.G. Gaustad, A.C. Boquest, B.E. Anderson, A.M. Gerdes, P. Collas, Differentiation of human adipose tissue stem cells using extracts of rat cardiomyocytes, Biochem. Biophys. Res. Commun. 314 (2004) 420–427. M. Brzoska, H. Geiger, S. Gauer, P. Baer, Epithelial differentiation of human adipose tissue-derived adult stem cells, Biochem. Biophys. Res. Commun. 330 (2005) 142–150. V. Planat-Benard, J.S. Silvestre, B. Cousin, M. Andre, M. Nibbelink, R. Tamarat, M. Clergue, C. Manneville, C. Saillan-Barreau, M. Duriez, A. Tedgui, B. Levy, L. Penicaud, L. Casteilla, Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives, Circulation 109 (2004) 656–663. Y. Cao, Z. Sun, L. Liao, Y. Meng, Q. Han, R.C. Zhao, Human adipose tissue-derived stem cells differentiate into endothelial cells in vitro and improve postnatal neovascularization in vivo, Biochem. Biophys. Res. Commun. 332 (2005) 370–379. K.M. Safford, K.C. Hicok, S.D. Safford, Y.D. Halvorsen, W.O. Wilkison, J.M. Gimble, H.E. Rice, Neurogenic differentiation of murine and human adipose-derived stromal cells, Biochem. Biophys. Res. Commun. 294 (2002) 371–379. A.I. Caplan, J.E. Dennis, Mesenchymal stem cells as trophic mediators, J. Cell Biochem. 98 (2006) 1076–1084. T.M. Liu, M. Martina, D.W. Hutmacher, J.H. Hui, E.H. Lee, B. Lim, Identification of common pathways mediating differentiation of bone marrow- and adipose tissue-derived human mesenchymal stem cells into three mesenchymal lineages, Stem Cells 25 (2007) 750–760. A.J. Katz, A. Tholpady, S.S. Tholpady, H. Shang, R.C. Ogle, Cell surface and transcriptional characterization of human adipose-derived adherent stromal (hADAS) cells, Stem Cells 23 (2005) 412–423. A.C. Boquest, A. Shahdadfar, K. Fronsdal, O. Sigurjonsson, S.H. Tunheim, P. Collas, J.E. Brinchmann, Isolation and transcription profiling of purified uncultured human stromal stem cells: alteration of gene expression after in vitro cell culture, Mol. Biol. Cell 16 (2005) 1131–1141. S. Gronthos, D.M. Franklin, H.A. Leddy, P.G. Robey, R.W. Storms, J.M. Gimble, Surface protein characterization of human adipose tissue-derived stromal cells, J. Cell Physiol. 189 (2001) 54–63. E. Zappia, S. Casazza, E. Pedemonte, F. Benvenuto, I. Bonanni, E. Gerdoni, D. Giunti, A. Ceravolo, F. Cazzanti, F. Frassoni, G. Mancardi, A. Uccelli, Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing T-cell anergy, Blood 106 (2005) 1755–1761. Y. Miyahara, N. Nagaya, M. Kataoka, B. Yanagawa, K. Tanaka, H. Hao, K. Ishino, H. Ishida, T. Shimizu, K. Kangawa, S. Sano, T. Okano, S. Kitamura, H. Mori, Monolayered mesenchymal stem cells repair scarred myocardium after myocardial infarction, Nat. Med. 12 (2006) 459–465. H. Ponta, L. Sherman, P.A. Herrlich, CD44: from adhesion molecules to signalling regulators, Nat. Rev., Mol. Cell Biol. 4 (2003) 33–45. D. Wainwright, L. Sherman, J. Sleeman, H. Ponta, P. Herrlich, A splice variant of CD44 expressed in the rat apical ectodermal ridge contributes to limb outgrowth, Ann. N.Y. Acad. Sci. 785 (1996) 345–349. R. Arch, K. Wirth, M. Hofmann, H. Ponta, S. Matzku, P. Herrlich, M. Zoller, Participation in normal immune responses of a metastasis-inducing splice variant of CD44, Science 257 (1992) 682–685. E XP E RI ME N TA L CE LL RE S EA RCH 3 14 ( 20 0 8 ) 6 0 3 –61 5 [18] U. Gunthert, M. Hofmann, W. Rudy, S. Reber, M. Zoller, I. Haussmann, S. Matzku, A. Wenzel, H. Ponta, P. Herrlich, A new variant of glycoprotein CD44 confers metastatic potential to rat carcinoma cells, Cell 65 (1991) 13–24. [19] D. Naor, R.V. Sionov, D. Ish-Shalom, CD44: structure, function, and association with the malignant process, Adv. Cancer Res. 71 (1997) 241–319. [20] C.R. Mackay, H.J. Terpe, R. Stauder, W.L. Marston, H. Stark, U. Gunthert, Expression and modulation of CD44 variant isoforms in humans, J. Cell Biol. 124 (1994) 71–82. [21] M. Krampera, A. Pasini, A. Rigo, M.T. Scupoli, C. Tecchio, G. Malpeli, A. Scarpa, F. Dazzi, G. Pizzolo, F. Vinante, HB-EGF/ HER-1 signaling in bone marrow mesenchymal stem cells: inducing cell expansion and reversibly preventing multilineage differentiation, Blood 106 (2005) 59–66. [22] M. Krampera, S. Glennie, J. Dyson, D. Scott, R. Laylor, E. Simpson, F. Dazzi, Bone marrow mesenchymal stem cells inhibit the response of naive and memory antigen-specific T cells to their cognate peptide, Blood 101 (2003) 3722–3729. [23] J.B. Mitchell, K. McIntosh, S. Zvonic, S. Garrett, Z.E. Floyd, A. Kloster, Y. Di Halvorsen, R.W. Storms, B. Goh, G. Kilroy, X. Wu, J.M. Gimble, Immunophenotype of human adipose-derived cells: temporal changes in stromal-associated and stem cell-associated markers, Stem Cells 24 (2006) 376–385. [24] K. McIntosh, S. Zvonic, S. Garrett, J.B. Mitchell, Z.E. Floyd, L. Hammill, A. Kloster, Y. Di Halvorsen, J.P. Ting, R.W. Storms, B. Goh, G. Kilroy, X. Wu, J.M. Gimble, The immunogenicity of human adipose-derived cells: temporal changes in vitro, Stem Cells 24 (2006) 1246–1253. [25] D.H. van Weering, P.D. Baas, J.L. Bos, A PCR-based method for the analysis of human CD44 splice products, PCR Methods Appl. 3 (1993) 100–106. [26] H. Castro-Malaspina, W. Ebell, S. Wang, Human bone marrow fibroblast colony-forming units (CFU-F), Prog. Clin. Biol. Res. 154 (1984) 209–236. [27] K. Yoshimura, T. Shigeura, D. Matsumoto, T. Sato, Y. Takaki, E. Aiba-Kojima, K. Sato, K. Inoue, T. Nagase, I. Koshima, K. Gonda, Characterization of freshly isolated and cultured cells derived from the fatty and fluid portions of liposuction aspirates, J. Cell Physiol. 208 (2006) 64–76. [28] S.J. Greco, K. Liu, P. Rameshwar, Functional similarities among genes regulated by OCT4 in human mesenchymal and embryonic stem cells, Stem Cells (in press). [29] R. Brandenberger, H. Wei, S. Zhang, S. Lei, J. Murage, G.J. Fisk, Y. Li, C. Xu, R. Fang, K. Guegler, M.S. Rao, R. Mandalam, J. Lebkowski, L.W. Stanton, Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation, Nat. Biotechnol. 22 (2004) 707–716. [30] D.J. Ceradini, A.R. Kulkarni, M.J. Callaghan, O.M. Tepper, N. Bastidas, M.E. Kleinman, J.M. Capla, R.D. Galiano, J.P. Levine, G.C. Gurtner, Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1, Nat. Med. 10 (2004) 858–864. 615 [31] K.E. McGrath, A.D. Koniski, K.M. Maltby, J.K. McGann, J. Palis, Embryonic expression and function of the chemokine SDF-1 and its receptor, CXCR4, Dev. Biol. 213 (1999) 442–456. [32] Y.R. Zou, A.H. Kottmann, M. Kuroda, I. Taniuchi, D.R. Littman, Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development, Nature 393 (1998) 595–599. [33] E. Raz, Guidance of primordial germ cell migration, Curr. Opin. Cell Biol. 16 (2004) 169–173. [34] C. Ries, V. Egea, M. Karow, H. Kolb, M. Jochum, P. Neth, MMP-2, MT1-MMP, and TIMP-2 are essential for the invasive capacity of human mesenchymal stem cells: differential regulation by inflammatory cytokines, Blood 109 (2007) 4055–4063. [35] M. Cheung, M.C. Chaboissier, A. Mynett, E. Hirst, A. Schedl, J. Briscoe, The transcriptional control of trunk neural crest induction, survival, and delamination, Dev. Cell 8 (2005) 179–192. [36] W.S. Wu, S. Heinrichs, D. Xu, S.P. Garrison, G.P. Zambetti, J.M. Adams, A.T. Look, Slug antagonizes p53-mediated apoptosis of hematopoietic progenitors by repressing puma, Cell 123 (2005) 641–653. [37] A. Inoue, M.G. Seidel, W. Wu, S. Kamizono, A.A. Ferrando, R.T. Bronson, H. Iwasaki, K. Akashi, A. Morimoto, J.K. Hitzler, T.I. Pestina, C.W. Jackson, R. Tanaka, M.J. Chong, P.J. McKinnon, T. Inukai, G.C. Grosveld, A.T. Look, Slug, a highly conserved zinc finger transcriptional repressor, protects hematopoietic progenitor cells from radiation-induced apoptosis in vivo, Cancer Cells 2 (2002) 279–288. [38] D. Rubio, J. Garcia-Castro, M.C. Martin, F.R. de la, J.C. Cigudosa, A.C. Lloyd, A. Bernad, Spontaneous human adult stem cell transformation, Cancer Res. 65 (2005) 3035–3039. [39] H. Zhu, N. Mitsuhashi, A. Klein, L.W. Barsky, K. Weinberg, M.L. Barr, A. Demetriou, G.D. Wu, The role of the hyaluronan receptor CD44 in mesenchymal stem cell migration in the extracellular matrix, Stem Cells 24 (2006) 928–935. [40] A. Bacigalupo, Mesenchymal stem cells and haematopoietic stem cell transplantation, Best. Pract. Res. Clin. Haematol. 17 (2004) 387–399. [41] K. Le Blanc, O. Ringden, Use of mesenchymal stem cells for the prevention of immune complications of hematopoietic stem cell transplantation, Haematologica 90 (2005) 438a. [42] P.V. Giannoudis, I. Pountos, Tissue regeneration. The past, the present and the future, Injury 36 (Suppl 4) (2005) S2–S5. [43] K. Le Blanc, I. Rasmusson, B. Sundberg, C. Gotherstrom, M. Hassan, M. Uzunel, O. Ringden, Treatment of severe acute graft-versus-host disease with third party haploidentical mesenchymal stem cells, Lancet 363 (2004) 1439–1441. [44] E. Zappia, S. Casazza, E. Pedemonte, F. Benvenuto, I. Bonanni, E. Gerdoni, D. Giunti, A. Ceravolo, F. Cazzanti, F. Frassoni, G. Mancardi, A. Uccelli, Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing T-cell anergy, Blood 106 (2005) 1755–1761.