Adipose tissue was obtained from healthy patients undergoing abdominal liposuction after written informed consent signed by the patients as approved by the French Ministry of Research and Innovation and the Personal data Protection ethics Committee (CPP) of Languedoc-Roussillon (approval AC-2010-1200).

Animal experimentation was conducted in agreement with the recommendations of Languedoc-Roussillon Regional Ethics Committee on Animal Experimentation. Approval of the protocol was given by the Languedoc-Roussillon Regional Ethics Committee.

Eight women were included in the study and the two techniques were used for each patient. The lipoaspirates were collected on both side of the umbilic in each patient. Liposuction of subcutaneous abdominal fat was carried out in patients under general anesthesia by an experienced surgeon.

The recovered fat tissues were transferred to 50 mL Falcon tubes and weighted. For enzymatical dissociation, an equal volume of 250 U/mL of collagenase type II (Ref C6885, Sigma, Saint-Quentin Fallavier, France) was added to the fat tissue and incubated at 37°C for 45 min under agitation as previously described (11). Cell suspensions were filtered using a sterile tea strainer and centrifuged at 300 × g for 10 min. The SVFs were collected, treated with erythrocyte lysis buffer (155 mmol/L NH₄Cl, 10 mmol/L KHCO₃, 0.11 mmol/L EDTA) and filtered successively through 100, 70, and 40 μm porous filters (Cell Strainer, BD Biosciences, le Pont-de-Claix, France). Cells were centrifuged at 300 × g for 5 min and resuspended in expansion medium (α-MEM medium supplemented with 100 U/mL penicillin/streptomycin, 2 mmol/mL glutamine, 10% fetal calf serum and 1 ng/mL bFGF) (R&D Systems, Abingdon, UK). Quantification of cells was performed using a hemocytometer and cells were plated at 4000 cells/cm² for 7 days. On day 7, cells were trypsinized, counted, and replated in expansion medium at the density of 2000 cells/cm² for another period of 7 days (end of passage 1).

Colony forming unit-fibroblast (CFU-F) efficiency was determined by seeding freshly isolated cells or ASCs at passage 1 at a concentration of 15 and 75 cells/cm² in 100-mm plates for 14 days. Cell colonies were fixed with methanol for 10 min, dried and stained with Giemsa solution for 45 min, rinsed with distilled H₂O, and let to dry. Colonies of more than 50 cells were counted.

Proliferation was measured as number of cell doubling per passage using the formulae: CD = ln(Nf/Ni)/ln2 where Nf is the final number of cells and Ni the initial number of cells. Population doubling time was calculated using the formulae: PDT = CT/CD; CT being the culture time.

Freshly isolated cells from the SVF and ASCs recovered at passage 1 (1 × 10⁶ cells) were suspended in phosphate buffered saline (PBS) containing 0.2% bovine serum albumin (BSA). Cells were then incubated with conjugated antibodies specific for CD11b, CD14, CD31, CD34, CD44, CD45, CD73, CD90, CD105, and HLA-DR and isotype-matched control antibodies (BD Biosciences) for 30 min on ice. Apoptotic cells were detected by FITC-conjugated Annexin V and 7AAD (BD Biosciences). Briefly, cells were washed twice with cold PBS, then suspended in 1X Annexin V binding buffer at 1 × 10⁶ cells/mL and incubated with FITC-Annexin V and 7AAD for 15 min at room temperature. Labeled cell acquisition was performed on a LSR Fortessa flow cytometer and data were analyzed using FACSDIVA 6.0 software (Becton Dickinson, San Jose, CA, USA).

For adipogenic differentiation, ASCs at passage 1 were plated at 8 × 10³ cells/cm² in expansion medium. At confluence (between days 3 and 5), the medium is replaced by the expansion medium (negative control) or the differentiation medium consisting in DMEM:F12 supplemented with 5% FCS, 100 U/mL penicillin (Lonza, Verviers, Belgium), 100 μg/mL streptomycin (Lonza), 16 μM biotin, 18 μM pantothenic acid, 100 μM ascorbic acid, 60 μM indomethacin, 450 μM IBMX, 1 μM dexamethasone, and 10⁻⁶ M rosiglitazone (all from Sigma). The medium was changed every 3 days for 21 days (from cell seeding). Differentiation was assessed by visualization of lipid droplets. Cells were fixed with 3% glutaraldehyde for 1 h, rinsed twice with H₂O and 60% isopropanol and then stained with Oil Red O dye (Sigma) for 2 h. Staining was then extracted by incubation with isopropanol for 10 min and quantified with a spectrophotometer at 540 nm.

For osteoblastic differentiation, ASCs at passage 1 were plated at 3 × 10³ cells/cm² in expansion medium. At confluence (between days 5 and 7), the expansion medium (negative control) or the differentiation medium was added. Osteogenic medium consists in DMEM containing 10% FCS, 2 mM glutamine (Lonza), 100 U/mL penicillin (Lonza), 100 μg/mL streptomycin (Lonza), 50 μg/mL ascorbic acid, 100 nM dexamethasone, and 3 mM NaH₂PO₄ (all from Sigma). The medium was changed every 3 days for 21 days (from cell seeding). Osteogenesis was assessed by visualization of the mineralized matrix after staining with 2% Alizarin Red S solution. Cells were rinsed with PBS, fixed with 95% ethanol for 30 min and incubated with Alizarin Red S solution for 5 min. Quantification of hydroxyapatite deposition was performed by incubation with 10% cetylpyridinium chloride for 10 min and the optical density (OD) was measured by a spectrophotometer at 562 nm.

Adipose stromal cells in proliferative medium were seeded in 96 wells plates at two densities (2 × 10⁴ or 4 × 10⁴ cells). Proliferative medium was IMDM supplemented with 2 mM l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, 0.1 mM non-essential amino acids, 0.25 mM β-2-mercaptoethanol, 1 mM sodium pyruvate, 10% inactivated FCS, and 20 mM Hepes. Peripheral blood mononuclear cells (PBMC) were isolated using Ficoll gradient from blood samples. PBMC were added in the wells containing the ASCs at the concentration of 2 × 10⁵cells/well and were activated with 2.5 μg/mL phytohemagglutinin. Cell cultures were incubated for 72 h. Proliferation of T lymphocytes was measured using the Cell Proliferation ELISA BrdU chemiluminescent kit (Roche Diagnostics, Meylan, France). Results are expressed as the percentage of proliferation of stimulated T cells and normalized to 100% for activated T cells.

BALB/c mice of 7 weeks old were immunized by injection of 100 μg chick ovalbumin (cOVA) emulsified in 100 μL complete Freund’s adjuvant at the base of the tail. After 5 days, delayed-type hypersensitivity (DTH) response was tested by challenging with 30 μg cOVA in 30 μL physiological saline solution in the right footpad. ASCs (2.5 × 10⁵ cells) were mixed with the cOVA solution before injection. The left footpad was injected with 30 μL saline solution as a negative control. At day 6, the thickness of the footpads was measured using a caliper and thickness increment was calculated as the difference between the immunized footpads at day 6 (right-left) and the unimmunized footpads at day 0 (right-left).

Statistical analysis was performed with GraphPad Software (San Diego, CA, USA). Comparison between groups used Shapiro–Wilk normality test. Data were compared using the Mann–Whitney’s test for non-parametric values. Results were expressed as the mean ± SEM (standard error of the mean); a p-value ≤0.05 was considered significant.

With the aim to compare the characteristics of ASCs isolated from fat tissue obtained after cannula- or water-jet-assisted procedures, we performed the two lipoaspiration procedures on the same patients. Eight women were included in the study and lipoaspirates from abdominal subcutaneous fat were withdrawn. The fat tissue obtained after water-jet-assisted lipoaspiration was generally less reddish than that obtained after the manual technique (Figure 1A). This was likely due to the high quantity of saline that was injected into the adipose tissue and discarded before fat tissue collection. After enzymatic dissociation of fat and SVF recovery, the cell yield was significantly different: 1.71 ± 0.81 and 4.29 ± 0.46 × 10⁵ cells/g tissue with the Bodyjet and manual procedure, respectively (Figure 1B). Viability of the cells was similar between the two procedures: 16.64 ± 2.72 and 12.9 ± 1.2% of AnnexinV⁺7AAD⁺ dead cells for the water-jet- or cannula-assisted technique, respectively (Figure 1C). We also determined the number of stem cells in the SVF suspensions by evaluating the number of CFU-F. The percentage of CFU-F in the SVF was identical in the two types of samples and estimated at 0.5 ± 0.2.5 and 0.6 ± 0.2% in the mononuclear cell fraction after Bodyjet and manual procedure, respectively (Figure 1D). Finally, we evaluated by fluorescence-activated cell sorting analysis whether the lipoaspiration procedures modified the phenotype of the SVF. Identical percentages of cell subtypes were detected in the two types of preparations, which contained 12–17% of CD31⁺ endothelial cells, 27% of CD45⁺ hematopoietic cells and 47% of ASC-containing CD34⁺ cells (Figure 1E). Other markers of hematopoietic subpopulations identified the presence of 2.5% of CD14⁺ monocytes, 15% of CD11b⁺ monocytes/macrophages, and 25% of HLA-DR⁺ cells. Evaluation of the markers of ASCs revealed the proportion of 68% of CD44⁺ cells, 40% of CD73⁺ cells, 50% of CD90⁺ cells, and 1.5% of CD105⁺ cells. More than 98% of the CD44⁺CD73⁺CD90⁺ cells were CD34⁺ confirming that ASCs were in the CD34⁺ cell population.

Stromal vascular fraction obtained after manual or water-jet-assisted procedures were plated at the density of 4000 cells/cm² and cultured to confluency for 7 days. After trypsination, reseeding and culture for 7 additional days, the adherent cells from the two types of techniques displayed similar elongated morphology with fibroblastic-like appearance (Figure 2A). Growth kinetics were identical between the two cell preparations. The cell number was multiplied by a 230-fold factor in 2 weeks (Figure 2B). Doubling time evaluation indicated similar values for water-jet-assisted or manual procedures: 54.5 ± 10.5 and 47.7 ± 9.4 h between days 0 and 7, respectively, and 40.9 ± 6.5 and 42.2 ± 7.2 h between days 7 and 14 (Figure 2C). The doubling time of ASCs isolated with the Bodyjet technique was significantly reduced between day 7 and day 14. We also determined the number of CFU-F in the adherent cell population at day 14. The same percentage of CFU-F was detected after expansion of cells isolated by the two procedures: 4% of the population was able to form colonies (Figure 2D). The viability of the cells after expansion was very good with 6% of dead cells and again no significant differences between the two techniques (Figure 2E). Finally, we analyzed the phenotype of cells after 14 days of culture. We noticed absence of endothelial and hematopoietic cells, as revealed by the lack of expression of specific markers for CD31, CD34, CD45, CD11b, and less than 1% CD14 (Figure 2F). By contrast, more than 99% of the cells expressed CD73, CD90, CD105, and CD44, while 15% of the cells expressed HLA-DR. Importantly, more than 99% of the CD44⁺, CD73⁺, CD90⁺, and CD105⁺ cells were in the CD34⁻ cell population (data not shown). Altogether, these results indicate that similar yield of ASCs was obtained upon culture, whatever was the procedure for fat isolation.

The differentiation potential of ASCs toward adipogenesis and osteoblastogenesis was investigated in vitro after 21 days of culture in inductive media. After adipogenesis induction, a high number of cells changed their morphology toward a rounder shape that accumulated lipid droplets (Figure 3A). In parallel, culture of the cells in expansion medium maintained the elongated fibroblastic morphology of the cells with no formation of lipid droplets. Lipid droplets were visualized after staining with Oil Red O. After extraction of the dye from the culture and spectrophotometric quantification, a highly significant increase of the OD measured at 540 nm was detected in the wells containing adipocytes as compared to the undifferentiated control (Figure 3B). The level of differentiation was, however, similar between ASC isolated from Bodyjet- or cannula-assisted techniques.

The potential to differentiate into osteoblasts was evaluated by culture in a medium containing a source of phosphate. Differentiation of ASCs and mineralization of the extracellular matrix were detected by Alizarin Red S staining. Whatever was the technique of fat isolation, ASC-derived osteoblasts highly mineralized the matrix as shown by red staining after culture in osteoinductive medium, while ASCs cultured in expansion medium were not able to differentiate (Figure 3C). Quantification of the dye after extraction nicely demonstrated a high staining in wells containing osteoblasts as compared to undifferentiated cells (Figure 3D). These data pointed out that the process for fat isolation did not impact the differentiation potential of ASCs.

We then evaluated in vitro the immunosuppressive capacity of ASCs. After 72 h culture of ASCs with PHA-activated PBMCs at different ratios, we measured the proliferative level of PBMCs. As compared to the proliferation of activated PBMC cultured alone, addition of ASCs led to a significant decrease of the PBMC proliferation (Figure 4A). Although not significant, the highest 1:5 MSC:PBMC ratio decreased slightly the proliferation of PBMC. Interestingly, when comparing the level of proliferation of PBMCs after culture with ASCs obtained from the two procedures of fat isolation, a significant improvement of the immunosuppressive effect of ASCs was detected with ASCs isolated after cannula-assisted technique (Figure 4A).

We, therefore, evaluated in vivo the immunomodulatory potential of six ASCs isolated from two different fat samples recovered by the two procedures in the DTH model. In this model, the immunization of mice with ovalbumin (OVA) at the basis of the tail followed by a boost in the fat pad after 5 days induced a high inflammatory reaction 24 h later, as visualized by the increase of the foot pad thickness (Figure 4B). The injection of ASCs together with OVA at the time of the boost reduced the inflammatory reaction and the foot pad thickness. Thickness of the foot pads was measured with a caliper and the average measures confirmed that ASCs were able to significantly reduce the inflammatory response (Figure 4C). However, there was no significant difference between ASCs isolated from fat recovered using the water-jet-assisted or manual procedure. Indeed, although ASCs obtained after manual liposuction were slightly more immunosuppressive in vitro, our in vivo data indicated that both techniques for fat sampling allowed isolating ASCs that are efficient to significantly reduce the host immune response.