Cells were obtained after informed consent from the resection of subcutaneous fat portions from healthy female donors who underwent breast plastic surgery (n = 6). SVF fraction was separated using a procedure modified from Zuk et al 511. Briefly, the tissue was digested with 0.075% collagenase in phosphate buffered-saline solution (PBS) at pH 7.4 for 45 min at 37°C (Cambrex Bio Science, Walkerville). Mature adipocytes and connective tissues were separated from the cell pellet by centrifugation at 800 ×g, for 10 min at 4°C. The cell pellet was resuspended in erythrocyte lysis buffer (155 mM NH₄Cl, 10 mM KHCO₃, 0.1 mM EDTA) and incubated for 10 min at room temperature in the dark. The cell suspension was then filtered through a 100 μm mesh filter (Becton Dickinson) and re-suspended in α-MEM/10%FBS. Nucleated cell count was performed both using trypan blue exclusion test and an automated cell counter (Sysmex K4500, Horger, CH). A portion of the cells was used for the Colony Forming Unit Fibroblast Assay (CFU-F). For the isolation and expansion of AT-MSC, the remaining cells of the SVF were plated in T25 flasks in α-MEM/10% FBS at a density of 100.000 cells/ml. This initial passage of the primary cell culture was referred to as passage 0 (P0). Cells were maintained in media until they achieved 75%–90% confluence. The cells were then replated at a density of 5.000 cells/cm² in T25 tissue culture flasks. The identity of AT-MSC cells was defined by using the criteria purposed by the ISCT committee 13: adherence to plastic, specific surface antigen expression and multipotent differentiation potential.

The CFU-F assay was used to evaluate the frequency of mesenchymal progenitors in the SVF 14. Briefly, cells from the SVF were resuspended in duplicate in 6-wells tissue culture plates at a final concentration of 10.000 cells/cm² in α-MEM/10%FBS and incubated at 37°C, 5% CO₂. After 14 days, the cells were washed with May-Grunwald Giemsa. Plates were scored under an inverted microscope (Nikon Eclipse TS100, Japan). Colonies were considered aggregates of more than 50 cells.

SVF cells were isolated by using the direct CD34 Progenitor Cell Isolation Kit (Miltenyi Biotec GmbH; Bergisch-Gladbach, Germany) following the manufacturer's protocol. Briefly, the SVF was incubated at 4°C for 30 min with the FcR blocking reagent and with magnetic microbeads coated with anti-CD34 primary antibody (QBEND/10). After washing, CD34+ cells were isolated by slow flow of cell suspension through a separation column placed in a magnetic field. To augment CD34+ cell purity, the separation step was repeated twice.

SVF cells (7 × 10⁴/slide) were spotted on cytospin slides by using a cytocentrifuge (Cytospin 3, Shandon Scientific Ltd, England), stained with May-Grunwald Giemsa and observed by using a Nikon Eclipse 50i microscope.

After primary culture in control medium (α-MEM/10% FBS + 50 U/ml Penicillin + 50 μg/ml Streptomycin, BioWhittaker, Walkersville, US), we examined the capacity of the cells to differentiate along osteogenic, adipogenic and chondrogenic lineage. Osteogenic induction: cells at passage 2 were plated at a density of 3.1 × 10³/cm² onto 2-wells chamber slides (BD Biosciences, Franklin Lakes, US). Cells were incubated in the control medium for one day to adhere them to the slides and the medium was replaced with Osteogenic Induction Medium from Osteogenic Differentiation Kit (Cambrex Bioscience, Ltd, Switzerland). Osteoinductive medium is a basal medium supplemented with dexamethasone, β-glycerol phosphate and ascorbic-acid-2-phospahate. On day 21 cells were fixed in 10% formalin for 30' and then osteogenic differentiation was assessed by incubating the fixed cells 10' in Alizarin Red S staining, 2% v/v in distilled water.

For the adipogenic induction, confluent cultures were incubated in Adipogenic Induction medium from Adipogenic Differentiation Kit (Cambrex Bioscience), while control cultures where fed with Adipogenic Maintenance medium from the same kit. Induction medium contains rh-Insulin, Dexamethasone, IBMX and Indomethacin, whereas maintenance medium contains only rh-Insulin. Induction was performed by culturing cells 3–4 days in induction medium, then 1–2 days in maintenance medium for 3 times. Adipogenesis was assessed by Oil O Red staining: cells fixed in 10% formalin were incubated in fresh Oil O Red water solution for 5'.(Fluka, Sigma-Aldrich, Saint Louis, US).

For the chondrogenic differentiation, AT-MSC were gently centrifuged in a 15-ml polypropylene conical tube to form small pellets and cultured 21 days in "differentiation basal medium" (Chondrogenic differentiation kit, Lonza) supplemented with 1 mM sodium pyruvate, 0.17 mM ascorbic acid-2-phosphate, 0.1 μM dexamethasone and 20 μg/ml TGF-β3. Every 3–4 days, cells were fed with fresh medium. Chondrogenic pellets were fixed in 10% formalin for 1 h at room temperature. Samples were the embedded in paraffin, sections stained with Alcian Blue and counterstained with Nuclear Fast Red.

Induced and uninduced cell layers were rinsed with D-PBS (BioWhittaker, Walkersville, US) after 21 days of adipogenic or osteogenic induction and immediately lysed using TRIzol (Invitrogen Corp. Carlsbad, CA, US). Total RNA was isolated using RNeasy Isolation Kit (Qiagen AG, Hombrechtikon, DE) and 1 μl was reverse-transcribed using Enhanced AMV Reverse Transcriptase and specific primers as indicated by the manufacturer (Sigma, Saint Louis, US). The resulting cDNA was, then, used as a template for PCR amplification with RedTaq Ready Mix (Sigma); the following genes have been inquired: glyceraldehydes-3 phosphate dehydrogenase (GAPDH) (NM_002046, F: 5'-TTCACCACCATGGAGAAGGC-3', R: 5'-GGCATGGACTGTGGTCATGA-3'), Osteocalcin (OSC), (NM_199173, F: 5'-CATGAGAGCCCTCACA-3', R: 5'-AGAGCGACACCCTAGAC-3'), Osteopontin(OSP), (NM_001040060, F: 5'-TTGCTTTTGCCTCCTAGGCA-3', R: 5'-GTGAAAACTTCGGTTGCTGG-3'), transcription factor PPARγ (NM_005037, F: 5'-TCAGGTTTGGGCGGATGC-3', R: 5'-TCAGCGGGAAGGACTTTATGTATG-3'), lipoprotein lipase (LPL) (NM_000237, F: 5'-GAGATTTCTCTGTATGGCACC-3', R: 5'-CTGCAAATGAGACACTTTCTC-3') and fatty acid binding protein 4 (FABP4) (NM_001442, F: 5'-ATGGGATGGAAAATCAACCA-3', R: 5'-GTGGAAGTGACGCCTTTCAT-3'). All primer sequences were original and determined through established mRNA GeneBank sequences. Glyceraldehydes-3-phosphate dehydrogenase was used as a control for assessing PCR efficiency. Reactions were performed in a T3000 thermal cycler (Biometra, Göttingen, Germany) for 35 cycles. Results were visualised by gel electrophoresis in 1.5% agarose (Promega, Madison, US) in a TAE buffer (Invitrogen) and stained with ethidium bromide (Promega). Reactions were repeated at least twice.

SVF cells were stained in quintuplicate with anti-human monoclonal antibodies against the following antigens: CD31 (clone 5.6E) and CD271 (clone ME20.4-1.H4) (FITC labeled), CD15 (clone VIMC6), CD73 (clone AD2), CD105 (clone 1G2) and CD133 (clone AC133/1) (PE labeled), CD45 (clone J33) (ECD labeled), CD13 (clone Immu103.44), CD14 (clone RMO52), CD146 (clone TEA 1/34), CD90 (clone Thy-1/310), (PC5 labeled) and CD34 (clone 581) (PC7 labeled). For the cell viability test, samples were incubated with 7-AAD. Isotype-matched murine FITC, PE, PC5, PC7 and ECD conjugated immunoglobulin were used as controls. All the antibodies were purchased by Beckman Coulter except for CD15, CD133 (Miltenyi Biotech, Germany) and CD73 (Becton Dickinson). Endothelial cells were further characterized by staining the cells with CD45, CD146 and 10 ug/ml of Ulex lecitin (UEA-1 FITC-labelled, Sigma Aldrich) for 1 h at room temperature in the dark. Huvec cells were used as a positive control.

A 5-color flow cytometric analysis was performed on a Cytomics FC500 cytofluorimeter (Beckman Coulter, Miami, FL) and data were analyzed with CXP software. Cells at passages 2 were analyzed for mesenchymal stem cells markers with the following anti-human monoclonal antibodies: CD31 (clone 5.6E), CD44 (clone J-174), CD29 (clone K20) and CD271 (clone ME20.4-1.H4) (FITC labeled), CD105 (clone 1G2), CD133 (clone AC133/1), CD166 (clone 3A6) and CD73 (PE labeled), CD45 (ECD labeled), CD38 (clone LS198-4-3), CD90 and CD13 (PC5 labeled) and CD34 (PC7 labeled). All Ab were from Beckman Coulter, except CD73 (Becton Dickinson), CD133 and CD271 (Miltenyi Biotec). Isotype-matched antibodies were also used.

The colony-forming cell haematopoietic assay (CFC) was performed on the crude SVF fraction and on enriched CD34+ cells after immunoselection from the SVF. Cells were cultured in Methocult GF H4434 (Stemcell Technologies Inc., Vancouver, Canada) containing 1% methylcellulose in Iscove's Modified Dulbecco's medium (IMDM), 30% Fetal Bovine Serum, 1% Bovine Serum Albumin, 10 × 10^-4 M 2-Mercaptoethanol, 2 mM L-glutamine, 3 U/ml recombinant human erythropoietin, 50 ng/ml rh Stem Cell Factor, 10 ng/ml rh GM-CSF and 10 ng/ml rh IL-3. Briefly, the SVF crude cells were plated in duplicate at a final concentration of 100.000 cells/ml and the SVF-CD34+ selected cells were plated at a final concentration of 10.000 cells/ml in tissue culture plates (Costar, Acton, MA). Plates were incubated at 37°C in a fully humidified atmosphere of 5% carbon dioxide and, after 14 days, each culture plate was examined under an inverted microscope. Granulocyte-erithrocyte-macrophage-megakariocyte colony-forming units (CFU-GEMM), granulocyte-macrophage colony-forming units (CFU-GM) and blast-forming units erythroid/mix (BFU-E/mix) were identified and counted using standard criteria. Colonies were considered aggregates of more than 50 cells.

The SVF consists of a heterogeneous cell population: based on the antigen expression, it is possible to individuate a subset of blood-derived cells. CD45 was detected with great variability on a mean percentage of 9.0 ± 6.9% of the SVF cells (range 1.5–17.4). The CD14, expressed on the monocytic lineage, was detected on 10.9 ± 9.6% of the cells (range 4.1–17.7) and CD13, expressed both on the early committed progenitors and mature granulocytes and monocytes was detected in 5.6 ± 3.9% of the cells (range 2.8–8.4). CD90, a marker associated with haematopoietic stem cells but also on connective tissue, was expressed on 29.2 ± 20.8% of the SVF (range 10.3–49.9). Complete data are reported in Table 1 and Table 2.

The percentage of SVF cells expressing CD34 has been reported with great variability among authors 11121516. In the present study, 6.9 ± 3.0% (range 3.4–10.2) of the SVF cells expressed the CD34 antigen. Haematopoietic progenitor cells are defined by the co-expression of the CD45+ cell marker: interestingly, when analyzing the co-expression of this antigen on CD34+ cells, only 2.0 ± 2.2% (range 0.4–3.5) of the CD34+ cells stained for CD45. 7.0 ± 5.9% (range 2.8–11) of the CD34+ cells expressed the CD90, a cell marker present on haematopoietic stem cells 17. The CD133 cell marker is co-expressed on a variable percentage of CD34+ haematopoietic progenitor cells, on stem cells 18, on developing epithelium and on the haemangioblast 19. In our cell populations, CD133 was expressed in only a small percentage of cells on the SVF (1.2 ± 1.5, range 0.40–3.80). To better characterize the antigenic profile of CD34+ cell subpopulation, we purified the CD34 cells by immunomagnetic selection. Cells were then subjected to FACS analysis and CFU haematopoietic assay. Two CD34+ cell subpopulations were present in the SVF fraction: a CD34 dim, accounting for about 90% of the CD34+ cells, and a CD34+ bright cell subpopulation (Figure 1). Among them, the CD45+ cells are mainly represented in the CD34 dim subpopulation (11.3%). The CD133+ is coexpressed by only 0.3% of the CD34+ cells. It is possible to conclude that only a small percentage of the SVF cells expressing the CD34+ antigen are haematopoietic progenitors.

Endothelial cell-associated markers, including CD31 20, CD105 21 and CD146 are expressed on SVF cells. The CD146 22, a cell marker expressed on mature endothelia, pericytes, endothelial progenitors cells (EPC) and at low intensity on a subset of activated T lymphocytes 23 stains 47.1 ± 3.2 of the cells (range 44.8–49.3) As can be observed in Figure 2, two distinct CD146 dim and CD146 bright cell clusters can be clearly distinguished. To finally confirm that mature endothelial cells are represented in the SVF, we have stained the cells in triplicate with CD45, CD146 and UEA-1. The results obtained suggest that it is possible to identify in the SVF a cell population having an endothelial phenotype defined as CD45-, CD146+ and UEA-1+ accounting for about the 7.7% of the total cells. (Figure 2).

Adherent cells generated after SVF expansion were observed after 4–7 days in culture. Cells rapidly generate a monolayer of fibroblastic-like morphology. At the second passage cells were analyzed and resulted to be uniformly negative (≤ 2%) for haematopoietic CD34, CD38, CD45, CD133, and non haematopoietic markers CD31 and CD271, and to be positive (≥ 95%) for CD13, CD29, CD44, CD73, CD90, CD105 and CD166 (Figure 3).

We evaluated both osteogenic, adipogenic and chondrogenic potential in AT-MSC at the second passage. For the osteogenic differentiation, morphological changes appeared during the second week of subculture. At the end of the 21-day induction period, some calcium crystals were clearly visible in the culture. Cell differentiation was confirmed by Alizarin Red staining. Osteogenic cell induction was confirmed at the molecular level: induced cells showed an upregulation of the expression of osteocalcin and osteopontin genes with RT-PCR assay.

The adipogenic potential was assessed by induction of confluent AT-MSC at second passage. At the end of the induction cycles, a consistent cell vacuolation was evident in the induced cells. Vacuoles brightly stained for fatty acid with Oil O Red staining. At the molecular level, induced cells showed differential expression of genes from the adipogenic pathway (LPL, FABP4 and PPARγ) (Figure 4).

Chondrogenic potential was assessed by induction of AT-MSC using the micromass method. Control cells did not even retain a spheroid shape and showed no specific staining while induced cells had a strong signal from Alcian Blue (Figure 5)

Morphologic analysis of the SVF revealed the presence of immature cells together with cells having a more differenciated morphology (Figure 6). May-Grunwald Giemsa staining (Figure 6, 100×).

The CFU-F assay was used to evaluate the frequency of mesenchymal progenitors in the SVF fraction. The estimated frequency of AT-MSC in the SVF was 1 in 4880 ± 1650 SVF cells (range 1826–6774).

To assess the haematopoietic potential of the SFV, the crude fraction (n = 5) and the CD34+ selected cells (n = 3) were plated in Methocult GF H4434. The frequency of haematopoietic colonies was comparable in the two cell fractions. Overall, the colonies were isolated with the following frequency (per 10⁵ cells): CFU-E, 3.3 ± 8.2 (range 0–20); BFU-E, 3.2 ± 6.0 (range 1–15); CFU-GM, 3.6 ± 8.1 (range 1–20); CFU-GEMM, 5.0 ± 12.3 (0–30). Images of haematopoietic colonies isolated from SVF are reported in Figure 7.