Effect of tissue-harvesting site on yield of stem cells derived from adipose tissue: implications for cell-based therapies
The stromal vascular fraction (SVF) of adipose tissue contains an abundant population of multipotent adipose-tissue-derived stem cells (ASCs) that possess the capacity to differentiate into cells of the mesodermal lineage in vitro. For cell-based therapies, an advantageous approach would be to harvest these SVF cells and give them back to the patient within a single surgical procedure, thereby avoiding lengthy and costly in vitro culturing steps. However, this requires SVF-isolates to contain sufficient ASCs capable of differentiating into the desired cell lineage. We have investigated whether the yield and function of ASCs are affected by the anatomical sites most frequently used for harvesting adipose tissue: the abdomen and hip/thigh region. The frequency of ASCs in the SVF of adipose tissue from the abdomen and hip/thigh region was determined in limiting dilution and colony-forming unit (CFU) assays. The capacity of these ASCs to differentiate into the chondrogenic and osteogenic pathways was investigated by quantitative real-time polymerase chain reaction and (immuno)histochemistry. A significant difference (P = 0.0009) was seen in ASC frequency but not in the absolute number of nucleated cells between adipose tissue harvested from the abdomen (5.1 ± 1.1%, mean ± SEM) and hip/thigh region (1.2 ± 0.7%). However, within the CFUs derived from both tissues, the frequency of CFUs having osteogenic differentiation potential was the same. When cultured, homogeneous cell populations were obtained with similar growth kinetics and phenotype. No differences were detected in differentiation capacity between ASCs from both tissue-harvesting sites. We conclude that the yield of ASCs, but not the total amount of nucleated cells per volume or the ASC proliferation and differentiation capacities, are dependent on the tissue-harvesting site. The abdomen seems to be preferable to the hip/thigh region for harvesting adipose tissue, in particular when considering SVF cells for stem-cell-based therapies in one-step surgical procedures for skeletal tissue engineering.
KeywordsAdipose-tissue-derived stem cells Chondrogenic differentiation Osteogenic differentiation Colony-forming unit assay Stem-cell yield Tissue-harvesting site Human
Tissue engineering is an emerging field in modern medicine. Therapies involve the combination of cells and scaffold materials that can be loaded with bioactive factors, ideally resulting in the regeneration or replacement of lost or damaged tissues and organs. Multiple cell sources have been investigated for their possible applicability in tissue engineering. Embryonal stem cells are the most potent stem cells; however, their use is controversial and has mayor ethical considerations (Dresser 2001). Mesenchymal stem cells (MSCs) can be obtained from the adult and are widely used because of their differentiation potential. In addition to bone marrow, periosteum (Nakahara et al. 1991), muscle (Asakura et al. 2001), and adipose tissue (Zuk et al. 2002) also appear to be sources of MSCs.
Subcutaneous adipose tissue is a particularly attractive reservoir of progenitor cells, because it is easily accessible, abundant, and self-replenishing. It is derived from the mesodermal germ layer and contains a supportive stromal vascular fraction (SVF) that can be readily isolated (Gronthos et al. 2003; Zuk et al. 2001). This SVF from adipose tissue consists of a heterogeneous mixture of cells, including endothelial cells, smooth muscle cells, pericytes, leukocytes, mast cells, and pre-adipocytes (Oedayrajsingh-Varma et al. 2006; Peterson et al. 2005; Prunet-Marcassus et al. 2005). In addition to these cells, the SVF contains an abundant population of multipotent adipose-tissue-derived stem cells (ASCs) that possess the capacity to differentiate into cells of mesodermal origin in vitro, e.g., adipocytes, chondrocytes, osteoblasts, and (cardio)myocytes (Erickson et al. 2002; Guilak et al. 2004; Halvorsen et al. 2001; Hattori et al. 2004; Planat-Benard et al. 2004; Rangappa et al. 2003; Zuk et al. 2001). Because of these favorable characteristics, interest has been growing in the application of ASCs for cell-based therapies such as tissue engineering.
For clinical practice, an advantageous approach would be to harvest ASCs and immediately give them back to the patient within the same operation, the so called “one-step surgical procedure” (Helder et al. 2007). This overcomes long-lasting culture expansion of ASCs on the one hand, but necessitates the use of the SVF of adipose tissue on the other hand, since fast selection procedures for stem cells in the SVF are not yet available. Therefore, the success of this procedure requires SVF-isolates to contain sufficient ASCs capable of differentiating into the desired cell lineage. In view of this, we have previously investigated the effect of three different surgical procedures for the harvesting of adipose tissue, i.e., resection, tumescent or conventional liposuction, and ultrasound-assisted liposuction, on the yield and function of the stem cells. We have demonstrated that the SVF isolates from adipose tissue harvested by ultrasound-assisted liposuction contain fewer stem cells, and that the stem cells have a longer population doubling time, leading us to conclude that resection and tumescent liposuction are preferable to ultrasound-assisted liposuction for harvesting adipose tissue, if the cells are to be used for tissue-engineering purposes (Oedayrajsingh-Varma et al. 2006).
In the present study, we have investigated whether the yield and functional characteristics of ASCs in the SVF are affected by the most frequently used adipose-tissue-harvesting sites. We have previously demonstrated that the yield of nucleated cells in the SVF of the adipose tissue from these different tissue-harvesting sites is similar (Oedayrajsingh-Varma et al. 2006). In the current study, the frequency of ASCs in the SVF cell isolates has been determined by using limiting dilution and colony-forming unit (CFU) assays. In addition, we have investigated the frequency of CFUs showing an osteogenic differentiation capacity. SVF cells have been subsequently cultured in order to obtain homogeneous cell populations and to acquire sufficient cells to determine their chondrogenic differentiation potential in micromass cultures. Homogeneity has been checked by determining the growth kinetics and phenotypic characteristics of the ASCs. To verify the maintenance of multidifferentiation potential, osteogenic and chondrogenic induction has been assessed in these homogeneous ASC cultures.
Materials and methods
Population characteristics (BMI body mass index, THS tissue-harvesting site, ASC adipose-derived stem cell, Res resection, T-LS tumescent liposuction, ND not determined)
ASC yield (%)
Cell isolation and storage
Isolation of the SVF from adipose tissue was performed as previously described (Oedayrajsingh-Varma et al. 2006). The isolation protocol included a Ficoll density centrifugation step to remove contaminating erythrocytes. After isolation, 4×106 SVF cells were resuspended in a mixture (1:1) of Dulbecco’s modified Eagle’s medium (DMEM) and cryoprotective medium (Freezing Medium, BioWhittaker, Cambrex, Verviers, Belgium), frozen under “controlled rate” conditions in a Kryosave (HCI Cryogenics, Hedel, The Netherlands), and stored in the vapor phase of liquid nitrogen according to standard practice at the Department of Pathology of the VU University Medical Center and following the guidelines of current Good Manufacturing Practice.
Limiting dilution assay
To assess the frequency of ASCs in the SVF of adipose tissue, SVF cells were seeded in normal culture medium, consisting of DMEM supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine (Invitrogen, Gibco, Calif., USA) in 96-well plates at 15×103 cells/well in the upper row. Two-fold dilution steps of the cells were made in subsequent rows. All cultures were performed in duplicate. Medium was changed twice a week. After 3 weeks, each well was individually scored for the number of cells. A well containing a cluster of at least 10 adhered fibroblast-like cells was considered as being positive. The frequency of ASCs was calculated from the rows of cells for which 25%–75% of the wells were scored as positive.
CFU assays were performed to check the consistency of the limiting dilution assay method and to determine the frequency of CFU capable of differentiating into the osteogenic lineage from the abdomen (n = 7) and hip/thigh region (n = 6). SVF cells were resuspended in normal culture medium, consisting of DMEM supplemented with 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine (Invitrogen, Gibco). Two 6-well plates were prepared in which the SVF was diluted ten-fold across both columns, resulting in a upper column containing 104 and a lower column containing 103 nucleated SVF cells.
For the CFU-fibroblast (CFU-F) assay, the fixation time was 11–14 days, depending on the amount and growth kinetics of the colonies (merging of colonies was avoided). At the appropriate time point, the medium was removed, and the cells were washed with phosphate-buffered saline (PBS), fixed with 4% formaldehyde for 10 min, and subsequently colored in a 0.2% toluidine blue solution in borax buffer for about 1 min. Excess stain was washed off with distilled water, and colonies were counted.
Cells of the duplicate 6-well plate were submitted to a CFU-alkaline phosphatase (CFU-ALP) assay. Cultures were performed in normal medium for 7 days in order to obtain colonies and to remove contaminating cells, after which osteogenic medium was added for 2 weeks. Following this period, cells in the CFU-ALP plate were rinsed with PBS, fixed in 4% formaldehyde, and incubated for 10 min in a 0.2 M TRIS-hydrochloride (pH 10), 0.2 M calcium chloride, 0.1 M magnesium chloride solution, whereafter a solution containing 0.2 M TRIS-hydrochloride (pH 10), 0.2 M calcium chloride, 0.1 M magnesium chloride, and 600 μl nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate was added for 30 min. The percentage of the colonies staining positive for ALP was determined.
Culturing of SVF cells
ASCs from the abdomen and hip/thigh region were cultured up to passage 2 for an adequate and quantitative comparison of stem-cell proliferation and differentiation capacity. Single-cell suspensions of cryopreserved SVF cells were seeded at 5.0×106 nucleated cells/cm2 in normal culture medium. The cultures were maintained in a 5% CO2 incubator at 37°C in a humidified atmosphere. The medium was changed twice a week. When reaching 80%–90% confluency, cells were detached with 0.5 mM EDTA/0.05% trypsin (Invitrogen) for 5 min at 37°C and replated. Cell viability was assessed by using the trypan blue exclusion assay. A homogeneous population of ASCs was thus obtained from abdomen and hip/thigh region and was subsequently checked by determining growth kinetics and by analyzing the surface-marker expression profile of the ASCs.
Growth kinetics of ASCs
Single-cell suspensions of cultured ASCs from abdomen and hip/thigh region were phenotypically characterized by using fluorescence-activated cell sorting (FACS; FACSCalibur, Becton Dickinson, USA) as previously described (Varma et al. 2007). All monoclonal antibodies (mAbs) were of the immunoglobulin G1 (IgG1) isotype. Cells were stained with fluorescently labeled antibodies (conjugated to fluorescein isothiocyanate, phycoerythrin, or allophycocyanin) against CD31, CD34, CD45, CD54, CD90, CD106, HLA-DR, and HLA-ABC (BD Biosciences, San José, Calif.), CD166 (RDI Research Diagnostics, Flanders, N.J.), CD105 (Caltag Laboratories, Burlingame, Calif.), CD117 (PharMingen, San Diego, Calif.), and CD146 (Chemicon, Temecula, Calif.). Nonspecific fluorescence was determined by incubating the cells with conjugated mAb anti-human IgG1 (DakoCytomation, Glostrup, Denmark).
Chondrogenic and osteogenic differentiation
The chondrogenic and osteogenic differentiation capacities of the cultured ASCs from the abdomen (n = 4) and hip/thigh region (n = 4) were studied. Chondrogenic differentiation was induced in cultured ASCs as previously described, with some modifications (Oedayrajsingh-Varma et al. 2006). In short, a 50-μl drop of a concentrated ASC cell suspension (8×106cells/ml, passage 2) was applied to a glass slide and allowed to attach at 37°C for 1 h. Then, 750 μl chondrogenic medium, consisting of DMEM, plus ITS+ Premix (final concentration in medium when diluted 1:100 was 6.25 μg/ml insulin, 6.25 μg/ml transferrin, 6.25 ng/ml selenous acid, 1.25 mg/ml bovine serum albumin (BSA), 5.35 μg/ml linoleic acid; BD, USA), 10 ng/nl transforming growth factor- β1 (TGF-β1; Biovision, ITK-diagnostics), 1% FCS, 25 μM ascorbate-2-phosphate (Sigma, St. Louis, Mo.), 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine, was overlaid gently. Cells were maintained in a 5% CO2/1% oxygen custom-designed hypoxia workstation (T.C.P.S. Rotselaar, Belgium) at 37°C in a humidified atmosphere, as this was shown to enhance chondrogenic differentiation (data not shown). Chondrogenic media were changed every 2–3 days.
For osteogenic differentiation, ASCs (passage 2) were seeded at 5000 cells/cm2 and cultured in monolayer in osteogenic medium, consisting of normal culture medium supplemented with 10 mM β-glycerol phosphate, 50 μg/ml ascorbate-2-phosphate, and 100 ng/ml bone morphogenetic protein 2 (BMP-2, Peprotech EC, London, UK). Osteogenic medium was changed twice a week.
(Immuno)histochemistry was performed as described previously (Oedayrajsingh-Varma et al. 2006). Cell nodules that formed under chondrogenic culture conditions were stained with Alcian blue (Sigma-Aldrich, Zwijndrecht, The Netherlands) at acidic pH for detection of proteoglycans. For the detection of collagen Type II, staining was performed with mouse monoclonal antibody II-II6B3 (1:50; Developmental Studies Hybridoma Bank, Iowa, USA) against human collagen Type II in PBS containing 1% BSA.
Osteogenesis was visualized after 21 days of culture in osteogenic medium by Von Kossa staining to establish the formation of a calcified matrix, typical for mature osteoblasts. The protocol used was as described previously (Varma et al. 2007) with only one modification of counterstaining the cytoplasm of the cells using fast green. Calcified extracellular matrix was visualized as black spots.
Spectrophotometric ALP activity
Early differentiation of MSCs into immature osteoblasts is characterized by ALP enzyme activity, with human MSCs expressing ALP as early as 4 days after induction, and maximum levels being observed at around 14 days after induction (Jaiswal et al. 1997). Therefore, cellular ALP activity was measured after culturing the ASCs in osteogenic medium for 14 days. Cells were lysed with distilled water, and the ALP activity and protein content were determined. To determine ALP activity, p-nitrophenyl phosphate (Merck, Darmstadt, Germany) at pH 10.3 was used as the substrate, as described by Lowry (1955). ALP activity was expressed as micromole per microgram of protein in the cell layer. The amount of protein was determined by using a BCA Protein Assay reagent Kit (Pierce, Rockford, Ill., USA), and the absorbance was read at 540 nm with a microplate reader (Biorad Laboratories, Hercules, Calif., USA).
Real-time polymerase chain reaction
RNA isolation and reverse transcription were performed as previously described (Oedayrajsingh-Varma et al. 2006). Real-time polymerase chain reactions (PCR) were performed by using the SYBRGreen reaction kit according to the manufacturer’s instructions (Roche Diagnostics) in a LightCycler 480 (Roche Diagnostics). cDNA (approximately 5 ng) was used in a volume of 20 μl PCR mix (LightCycler DNA Master Fast Startplus Kit, Roche Diagnostics) containing a final concentration of 0.5 pmol primers. Relative housekeeping gene expression for 18 S-rRNA (18 S) and relative target gene expression for aggrecan (AGG), collagen Type II (COL2B), and collagen Type X (COL10α1) regarding chondrogenic differentiation, and for collagen Type I (COL1α), osteopontin (OPN), and runt-related transcription factor 2 (RUNX-2) regarding osteogenic differentiation were determined.
Primers (Invitrogen) used for real-time PCR are listed in Table 2. They were designed by using Clone Manager Suite software program version 6 (Scientific & Educational Software, Cary, N.C., USA). The amplified PCR fragment extended over at least one exon border, based on homology in conserved domains between human, mouse, rat, dog, and cow, except for the 18 S gene (encoded by one exon only). Amplified Col2B PCR products were electrophoresed on a 2% agarose gel and stained with ethidium bromide. For real-time PCR, the values of relative target gene expression were normalized to relative 18 S housekeeping gene expression.
PCR primer sets used for reverse transcription/PCR (18 S 18 S subunit, AGG aggrecan, COL1αI collagen Type I, COL2B collagen Type II, COL10a1 collagen Type X, OPN osteopontin, RUNX-2 runt-related transcription factor 2)
Accession number, product length (bp)
Forward: 5′ GTAACCCGTTGAACCCCATT− 3′
Human, NM_10098, 151 bp
Reverse: 5′ CCATCCAATCGGTAGTAGCG 3′
Forward: 5′CAACTACCCGGCCATCC 3′
Human, NM_001135, 160 bp
Reverse: 5′GATGGCTCTGTAATGGAACAC 3′
Forward: 5′ AAGCCGAATTCCTGGTCT 3′
Human, NM_000088, 195 bp
Reverse: 5′ TCCAACGAGATCGAGATCC 3′
Forward: 5′ AGGGCCAGGATGTCCGGCA 3′
Human, NM_033150, 195 bp
Reverse: 5′ GGGTCCCAGGTTCTCCATCT 3′
Forward: 5′ CACTACCCAACACCAAGACA 3′
Human, NM_000493, 225 bp
Reverse: 5′ CTGGTTTCCCTACAGCTGAT′
Forward: 5′ TTCCAAGTAAGTCCAACGAAAG 3′
Human, AF_052124, 181 bp
Reverse: 5′ GTGACCAGTTCATGAGATTCAT 3′
Forward: 5′ ATGCTTCATTCGCCTCAC 3′
Human, NM_001024630, 156 bp
Reverse: 5′ ACTGCTTGCAGCCTTAAAT 3′
Real-time PCR data analysis
With the Light Cycler software (version 4), the crossing points were assessed and plotted versus the serial dilution of known concentrations of the standards derived from each gene by the Fit Points method. PCR efficiency was calculated by Light Cycler software, and the data were used only if the calculated PCR efficiency was between 1.85–2.0.
Kolmogorov-Smirnov tests were used to determine the normalcy of measurements and, if appropriate, their logarithmics. For the evaluation of yield and growth kinetics, means between two groups in one variable were compared by using the independent sample two-tailed t-test. Partial correlation was expressed as the Pearson correlation coefficient, r. For evaluation of gene expression, a repeated measures analysis of variance was used to determine significant differences when increasing time-points in one donor within one variable were compared. If levels of gene expression were below the detection limit (0.05), values were set at 10−2 (or log level at −2). All statistical tests used a significance level of α = 0.05.
Effects of tissue-harvesting site on frequency of ASCs
Effect of tissue-harvesting site on frequency of CFUs having osteogenic differentiation potential
Phenotypic characterization and growth kinetics of cultured ASCs
Surface-marker expression of human cultured ASC at passages 3–4. Results are expressed as mean fluorescence (MF), with isotype control <4.25 (− MF<8.5; + 8.5<MF<100; ++ 100<MF<1000; +++ MF>1000).
Cultured ASC (n = 4)
Effect of tissue-harvesting site on osteogenic differentiation potential of ASCs
ALP activity in the ASCs was measured after 14 days of osteogenic stimulation. ASCs cultured in control medium served as negative controls. ALP activity in the osteogenically stimulated cells was significantly increased (P = 0.047) compared with that in control cells (Fig. 4d). No statistically significant difference was apparent in ALP activity between ASCs derived from abdominal fat and ASCs from hip/thigh fat. Calcification of the osteogenic matrix was confirmed by using Von Kossa staining; black spots could be observed after 3 weeks of osteogenic stimulation of ASCs from both origins (Fig. 4e: abdomen, Fig. 4f: hip/thigh). No calcification was seen in ASC cultures expanded in control medium (Fig. 4g).
Effect of tissue-harvesting site on chondrogenic differentiation potential of ASCs
The chondrogenic differentiation potential of ASCs was analyzed after culturing the cells in a micromass in chondrogenic medium containing TGF-β. Within 24 h of culture, most of the cells formed nodules (n = 7).
Alcian blue staining demonstrated proteoglycan deposition in the ASC nodules from both the hip/thigh region and abdomen (Fig. 5d). At higher magnification (Fig. 5e), the ASC nodules resembled cartilage-like tissue, composed of round cells, surrounded by lacunae and lying in a proteoglycan-rich extracellular matrix that appeared positive for collagen Type II by immunostaining (Fig. 5f).
In this study, we have investigated whether the yield and functional characteristics of ASCs are affected by the adipose tissue-harvesting site, i.e., abdomen and hip/thigh regions. We have found a difference in the frequency of ASCs between adipose tissue harvested from the abdomen and the hip/thigh regions. SVF isolates derived from abdominal fat contain significantly higher frequencies of ASCs. When cultured, the growth kinetics and surface-marker expression of ASCs from both tissue-harvesting sites are similar. We have detected no differences in osteogenic or chondrogenic differentiation potential between these cultured ASCs from the two tissue-harvesting sites.
Adipose tissue is a highly heterogeneous tissue, not only among individuals, but also when comparing different fat depots within one individual. Donor-dependent differences have been demonstrated to exist in (stem) cell yield, proliferation, and differentiation capacity, probably caused by differences in age (Hauner and Entenmann 1991; van Harmelen et al. 2003), BMI (Aust et al. 2004; Hauner et al. 1988; Jaiswal et al. 1997; van de Venter et al. 1994), and diseases such as osteoarthritis and diabetes (Barry 2003; Murphy et al. 2002; Ramsay et al. 1995). Donors used in this study for harvesting adipose tissue were healthy female donors up to 62 years old. Some patients who will benefit from tissue engineering of cartilage will be older or might suffer from disease, and males will also be affected. Therefore, future research should include these donor types to determine whether yields and functional characteristics are influenced by these variables.
In our donor population, no significant correlation has been detected between the frequency of ASCs and the age of the donor (P = 0.32, r = 0.27) or between the frequency of ASCs and BMI (P = 0.42, r = −0.22; data not shown). These data are in agreement with other studies that have demonstrated no correlation between BMI or age and numbers of ASCs per gram of adipose tissue (Hauner and Entenmann 1991; van Harmelen et al. 2003). Most importantly, this means that these variables cannot be responsible for the difference that we have found between the frequency of ASCs and the tissue-harvesting site.
In addition to donor-dependent heterogeneity, intra-individual differences between fat depots have been demonstrated, e.g., with regard to the metabolic response of adipose tissue to various hormonal and neurological stimuli (Guilak et al. 2004; Lacasa et al. 1997; Masuzaki et al. 1995; Monjo et al. 2003; Rodriguez-Cuenca et al. 2005) and the cellular composition of the adipose tissue (Peptan et al. 2006; Prunet-Marcassus et al. 2005). In this study, we have focused on the intra-individual differences in the yield and function of ASCs from two fat depots: the abdomen and the hip/thigh region. We have demonstrated that adipose tissue derived from the abdomen contains significant higher frequencies of stem cells compared with adipose from the hip/thigh region. Whereas SVF isolates from the abdomen in this study contain about 5.1% of ASCs, the frequency of ASCs in SVF from the hip/thigh region is only 1.2%. Moreover, this 1.2% is the average of a population in which some donors hardly possess any adipose-derived stem cells (see spreading ASC frequency hip/thigh in Fig. 1). Despite this more than four-fold difference, the frequency of ASCs in adipose tissue from the hip/thigh region is still much higher compared with the frequency of MSCs in the bone marrow compartment, which is as low as 0.001%–0.01% (Pittenger et al. 2000), thereby making the hip/thigh region still a much more attractive stem-cell source for tissue engineering therapies.
What are the implications of these stem-cell frequencies for clinical practice? We have shown that 0.5–2.0×108 SVF cells can be harvested from 100 g adipose tissue, an amount that can easily be obtained from a patient (Aust et al. 2004; De Ugarte et al. 2003; Oedayrajsingh-Varma et al. 2006; Zuk et al. 2001). With an ASC frequency of 5.1%, the SVF isolates contain between 2.6–10.2×106 stem cells, which is an amount that appears to be sufficient for cell-based therapies as compared with the amount of cells used by others (Erickson et al. 2002; Fan et al. 2006; Williams et al. 2003; Zheng et al. 2006). This implies that time-consuming culturing and expanding steps of the stem cells can be avoided. In comparison, a bone marrow transplant of 100 ml contains approximately 6.0×108 nucleated cells (Zuk et al. 2002), of which only 0.001%–0.01% (0.06–0.006×106 cells) are stem cells (Pittenger et al. 2000).
To determine the frequency of CFU capable of differentiating into the osteogenic lineage, a CFU-ALP assay has been performed on cells from the abdomen and hip/thigh region. The frequencies of CFU-ALP ( ± 3%) are almost comparable with those of the CFU-F (Figs. 1b, 2). This is higher than the value that Mitchell et al. (2006) have found in their clonogenic assay of freshly isolated stromal cells (±0.5%). However, they have used a different assay method and shorter incubation period, and their frequency increases to the same level (5%) after progressive passaging (Mitchell et al. 2006).
SVF cells have been cultured up to passage 3 to obtain a homogeneous population of ASCs as starting material for differentiation studies toward the chondrogenic lineage. When comparing the growth kinetics of these cultured ASCs from the abdomen and hip/thigh region, the cell doubling time appears to be similar, being approximately 2 days. This is in accordance with the findings of others who have compared the replication rate of adipocyte precursor cells from various tissue-harvesting sites (Hauner et al. 1988; Pettersson et al. 1985; Roncari et al. 1981; Zuk et al. 2001). On the other hand, when comparing SVF cells from omental and subcutaneous fat, van Harmelen et al. (2004) have found a difference in cell proliferation rate; however, this may be explained by differences in methodological approach, since they use stromal cells instead of cultured ASCs. As we have shown that different fat depots contain different numbers of stem cells, these differences in proliferation rate may be caused by differences in initial stem-cell numbers when using SVF cells. This is reflected in the finding that, in our study, SVF cells derived from abdominal fat reach 80%–90% confluency within 5 days, whereas SVF cells derived from adipose tissue of the hip/thigh region take more than 9 days to reach 80%–90% confluency when seeded in the same density (data not shown).
In addition to the determination of growth kinetics, we have phenotypically characterized the cultured ASCs. The surface-marker expression profile is in accordance with those found by others (Mitchell et al. 2006; Schaffler and Buchler 2007; Varma et al. 2007), making both tissue sites fully comparable with each other (Tables 1, 3).
The homogeneous ASC population has been induced to the osteogenic and chondrogenic lineages. Having determined the osteogenic differentiation capacity of ASCs, we have shown significant up-regulation of osteogenic gene expression, ALP activity, and matrix mineralization. Interestingly, although no significant difference has been detected in ALP activity between ASCs from the abdomen and hip/thigh regions, ASCs from the hip/thigh region tend to show higher values of ALP activity after induction. This might be related to the underlying bone tissue, thereby implying that the ASCs of the hip/thigh region are less multipotent and more committed to the osteogenic lineage.
The chondrogenic differentiation capacity of the ASCs has been demonstrated by the up-regulation of AGG and COL10α1 gene expression and the production of matrix proteins. No difference has been detected in the chondrogenic differentiation potential between ASCs from the abdomen and hip/thigh regions. However, in the PCR studies, we have not succeeded in quantifying COL2α and Col2B mRNA expression for any of the donors tested. Others have obviously faced the same difficulty when trying to measure the up-regulation of COL2B genes in MSCs (Huang et al. 2004; Winter et al. 2003; Zuk et al. 2002). Although intending to measure collagen Type X mRNA expression as a hypertrophic and therefore late marker of chondrogenesis, we have noticed that this mRNA is expressed earlier than collagen Type II mRNA. This is surprising, as we would expect stem cells to have to differentiate into chondrocytes before they can become hypertrophic. However, this unexpected hierarchy of chondrogenic gene expression has also been found by Mwale et al. (2006). Moreover, both Mwale et al. (2006) and we have found AGG to be constitutively expressed in MSCs. Because of this constitutive expression of AGG and the early up-regulation of COL10α1, Mwale et al. (2006) warn against using these molecules as markers for chondrogenesis and chondrocytic hypertrophy. We think that these genes can nevertheless be used as markers for differentiation into the chondrogenic lineage, albeit being exclusively shown as the quantitative up-regulation of gene expression, and always in combination with other chondrogenic markers (an awareness of the possible difference in the function of COL10α1 in chondrogenesis in adult stem cells when compared with embryonic stem cells is also necessary).
Our lack of detection of any significant differences in the osteogenic and chondrogenic differentiation potential when comparing ASCs from the two tissue-harvesting sites seems to be in contrast with studies of Hauner and Entenmann 1991) who have found differences in the adipogenic differentiation potential between SVF cells from abdominal and femoral adipose tissue. However, since Hauner and Entenmann 1991) have used fresh SVF cells instead of culture-passaged ASCs, the variation in differentiation potential might be attributable to differences in numbers of ASCs in the SVF isolates of the two regions, as we have shown in this study. Other factors responsible for the variation in differentiation potential can be ascribed to other specific histological characteristics of the adipose tissue at the anatomical site, such as vascularity and amount of fibrous tissue (Lennon et al. 2000; Peptan et al. 2006; Pittenger et al. 2000), and to differences in the regulation of gene expression (Djian et al. 1983; Peptan et al. 2006; Pittenger et al. 2000).
We therefore conclude that the yield of ASCs is dependent on the tissue-harvesting site. In planning the optimal one-stage procedure for the regeneration of cartilage tissue, factors that can positively influence the outcome of the operation must be taken into account. In view of this, the abdomen seems to be preferable to the hip/thigh region for harvesting ASCs.
The authors thank the Jan van Goyen Clinic for providing adipose tissue, the Department of Physiology for use of their hypoxia workstation, Zufu Lu for his technical assistance with PCR analyses, and Jolanda de Blieck for her assistance in the lab.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
- Dresser R (2001) Ethical issues in embryonic stem cell research. JAMA 285:1439–1440Google Scholar
- Lowry OH (1955) Micromethods for the assay of enzyme. II Specific procedure. Alkaline phosphatase. Methods Enzymol 4:371–372Google Scholar
- Mitchell JB, McIntosh K, Zvonic S, Garrett S, Floyd ZE, Kloster A, Di Halvorsen Y, Storms RW, Goh B, Kilroy G, Wu X, Gimble JM (2006) Immunophenotype of human adipose-derived cells: temporal changes in stromal-associated and stem cell-associated markers. Stem Cells 24:376–385PubMedCrossRefGoogle Scholar
- Planat-Benard V, Silvestre JS, Cousin B, Andre M, Nibbelink M, Tamarat R, Clergue M, Manneville C, Saillan-Barreau C, Duriez M, Tedgui A, Levy B, Penicaud L, Casteilla L (2004) Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives. Circulation 109:656–663PubMedCrossRefGoogle Scholar
- Winter A, Breit S, Parsch D, Benz K, Steck E, Hauner H, Weber RM, Ewerbeck V, Richter W (2003) Cartilage-like gene expression in differentiated human stem cell spheroids: a comparison of bone marrow-derived and adipose tissue-derived stromal cells. Arthritis Rheum 48:418–429PubMedCrossRefGoogle Scholar