Human Serum from Platelet-Poor Plasma for the Culture of Primary Human Preadipocytes


  • Eva Koellensperger,

    1. Department of Plastic Surgery and Hand Surgery, Burns Center, University Hospital, Aachen University of Technology, Aachen, Germany
    Search for more papers by this author
  • Dennis von Heimburg,

    1. Department of Plastic Surgery and Hand Surgery, Burns Center, University Hospital, Aachen University of Technology, Aachen, Germany
    Search for more papers by this author
  • Marta Markowicz,

    1. Department of Plastic Surgery and Hand Surgery, Burns Center, University Hospital, Aachen University of Technology, Aachen, Germany
    Search for more papers by this author
  • Norbert Pallua M.D.

    Corresponding author
    1. Department of Plastic Surgery and Hand Surgery, Burns Center, University Hospital, Aachen University of Technology, Aachen, Germany
    • Department of Plastic and Hand Surgery, University of Heidelberg, Ludwig-Guttmannstrasse 13, 67071 Ludwigshafen, Germany. Telephone: +49-621-6810-0; Fax: +49-621-6810-2327
    Search for more papers by this author


In adipose tissue engineering, the use of human serum is essential to achieve the goal of an autologous system. Serum from conventional human plasma (SCP) contains platelet-derived growth factor (PDGF), a growth factor known to be both a potent inhibitor of adipose differentiation and also the most important stimulator of proliferation in human serum. Serum from platelet-poor plasma (SPPP) is considered to be PDGF-deprived and should therefore inhibit the differentiation of preadipocytes to adipocytes to a lesser extent. Effective cultivation of preadipocytes with SPPP requires compensating for the missing stimulatory PDGF effect on proliferation. However, the addition of other growth factors to the media needs to provide stimulation of proliferation without significant inhibition of differentiation. Primary human preadipocytes were isolated from adipose tissue samples of 10 healthy human donors and cultured under four different medium conditions (SCP, SPPP, SPPP + 1 nM basic fibroblast growth factor [bFGF], and SPPP + 1 nM epidermal growth factor [EGF]) for five generations. Proliferation activity and differentiation capacity were assessed for each sample, generation, and culture condition by calculating doubling time and measuring glycerol-3-phosphate dehydrogenase (GPDH)-specific activity. The use of SPPP resulted in a marked rise in GPDH activity compared with the cells cultured with SCP. Supplementing SPPP with 1 nM bFGF or EGF increased proliferation activity significantly. SPPP can be considered superior to SCP for the culture of primary human preadipocytes in adipose tissue engineering in terms of proliferation activity and differentiation capacity.


Human serum contains a wealth of different growth factors and mitogens, such as platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), vascular endothelial-derived growth factor, or insulin-like growth factor (IGF-I) [18]. PDGF is a major mitogen, especially for connective tissue-like cell types; it also inhibits apoptosis. In the adult, it is important in wound healing. During embryogenesis, it plays an important role in the development of the central nervous system, kidneys, blood vessels, and lungs. As a dimeric molecule, it consists of disulfide-bonded A- and B-polypeptide chains, which combine to homo- and heterodimers. A major store of PDGF in humans is the α-granules of blood platelets [9, 10]. Total PDGF activity in serum differs more than 15-fold among different species, from less than 1 ng/ml in dog, chicken, pig, and calf to greater than 13 ng/ml in mouse and human. However, as the total mitogenic activity of these sera is comparable [2], the relative importance as serum mitogen varies considerably. Human platelets contain all three different PDGF isoforms, mainly PDGF AB, with a median serum level of 23 ng/ml, but they also contain small amounts of PDGF BB (median serum level 0.32 ng/ml) and AA [1, 8, 11]. In addition, they contain substantial amounts of EGF. After coagulation, with EGF being released from platelets, conventional serum contains levels in the range of 0.5–2 ng/ml, sufficient to induce mitosis and cell migration [7, 12].

Serum from platelet-poor plasma (SPPP) is a special form of plasma almost entirely free of platelets and PDGF [13]. In general, platelets and erythrocytes are removed from blood by centrifuging [14]. When conventional serum is obtained from plasma, PDGF is secreted from aggregating platelets during coagulation. Because the platelet count is low, less PDGF is released into the serum when platelet-poor plasma coagulates. To our knowledge, the use of serum from platelet-poor plasma for the culture of primary human preadipocytes has not been previously described.

In this study, we tried to minimize the known inhibiting effects of PDGF on the differentiation of primary human preadipocytes [15, 16] by using PDGF-poor serum from platelet-poor plasma. The proliferative properties of PDGF were to be substituted by the addition of other growth factors, namely bFGF and EGF. The aim was to find the growth factor that would best stimulate proliferation while having the least inhibiting effect on differentiation. bFGF and EGF were chosen to compensate for the missing stimulatory effect of PDGF on proliferation, because previous experiments suggested that they might act as stimulators of both preadipocyte proliferation and differentiation [15, 1719].

First isolated by Gospodarowicz et al. [20], bFGF promotes wound healing, tissue repair processes, hematopoiesis, angio-genesis, and mitogenic activities in cells of epithelial, mesenchymal, and neuronal origin, as well as in human embryonic stem cells via a wealth of different fibroblast growth factor (FGF) receptors [2127]. bFGF release from cells is thought to be associated with cell lysis and damage of the plasma membrane caused by cell death, wound injury, chemical injury, irradiation, or infection [22]. Released bFGF is stored in the extracellular matrix and basement membranes bound to heparan sulfate proteoglycans, where its distribution is controlled by diffusion with rapid reversible binding [2830]. The median bFGF level in normal human serum is approximately 0.7 pg/ml [8].

EGF, a polypeptide with two sets of antiparallel β-sheet structures, has an impact on proliferation, differentiation, and many other aspects of a range of mammalian cells, including corneal and mammary epithelium and skin cells [31, 32]. EGF promotes wound healing by stimulating migration and division of epithelial cells, keratinocytes, and vascular endothelial cells and by increasing protein synthesis. Fibroblast cell number is enhanced through chemotaxis and mitosis, resulting in higher collagen production. Overactive signaling through the EGF system also plays a role in tumor growth. EGF acts via a trans-membrane glycoprotein receptor with its cytoplasmatic tyrosine kinase domain through phosphorylation of several important regulatory proteins. One important result is the transcription of early genes required for mitosis (e.g., c-jun) [1, 33].

The differentiation of preadipocytes into adipocytes is influenced by multiple different factors. This process, also called adipogenesis, can be quantified biochemically via glycerol-3-phosphate dehydrogenase activity (GPDH). GPDH is a specific marker of adipogenesis and catalyses the formation of glycerol-3-phosphate, an important step in the synthesis of triacylglycerines [34]. The most common in vitro models to analyze the effects of different growth factors, mitogens, and medium components on adipose differentiation are 3T3-L1 and 3T3-F442A cell lines and murine Swiss 3T3-cells [35].

In serum-containing medium, the addition of corticosterone, IBMX, and 1 μM insulin is sufficient to trigger complete adipocyte differentiation of 3T3-L1 preadipocytes. Adipose conversion of 3T3-L1 cells in a serum-free culture system, however, depends on corticosterone, cyclic AMP, and the addition of growth factors, such as PDGF, bFGF, IGF-1, or EGF [3638]. Bachmeier et al. have shown that in 3T3-L1 preadipocytes grown to confluence under serum-free conditions, PDGF BB, as well as EGF, maximally stimulated adipocyte differentiation at a concentration of 1 nM, whereas bFGF was less potent [37, 38]. At higher concentrations, GPDH activity decreased, whereas cell proliferation was further stimulated. Specific GPDH activities obtained in 3T3-L1 preadipocyte differentiation in a serum-free system with supplementation of either PDGF BB or EGF were in the same range as in a culture system containing serum. Partially in contrast to these results, Krieger-Brauer et al. have shown that different members of the fibroblast and platelet-derived growth factor families can induce antagonistic effects on adipose conversion in 3T3 L1-cells cultured with 5% fetal calf serum: bFGF and PDGF BB inhibited the expression of the adipocyte phenotype induced by insulin, dexamethasone, and IBMX, whereas acidic fibroblast growth factor and PDGF AA stimulated adipose conversion [39].

However, results obtained from cell lines need to be interpreted with care when transferred to primary cells, as there are a few differences, especially regarding stimulation of proliferation and differentiation (e.g., primary cells have a limited life span, and their differentiation capacity markedly depends on age, gender, site, and species) [35, 4042].

Materials and Methods

Tissues and Cells

Preadipocytes were isolated from the freshly excised human subcutaneous adipose tissue of 10 healthy adults (two males and eight females in the age range 24–62, with a median age of 37.7 years) undergoing elective plastic surgery procedures at the Department of Plastic Surgery and Hand Surgery, Burns Center (Aachen University of Technology, Aachen, Germany). None of the patients had diabetes, had a severe systemic illness, or was taking medications known to affect adipose tissue metabolism.

Isolation of Preadipocytes

Following the removal of fibrous tissue, the adipose tissue samples were minced and digested enzymatically by collagenase (collagenase type CLS, 196 U/mg; Biochrom AG, Berlin, at 37°C for 45 minutes under constant shaking. After filtration through a 250-μm nylon mesh (Neolab, Heidelberg, Germany,, the digestion process was stopped by adding M199 (powdered M199; Biochrom AG), 10% fetal calf serum (Biochrom AG), 100 U/ml penicillin (PAA Laboratories, Linz, Austria,, and 0.1 mg/ml of streptomycin (PAA Laboratories). The cell suspension was centrifuged at 700g and room temperature for 7 minutes, and the pellet was resuspended in M199, 10% fetal calf serum, and 100 U/ml penicillin-0.1 mg/ml streptomycin. The cells were equally plated into four sterile tissue culture flasks at an average density of 20,000 cells per cm2 and maintained at 37°C in an atmosphere of 5% CO2.

Cell Culture

SCP and SPPP were purchased from Sigma-Aldrich (St. Louis, The concentrations of bFGF and EGF were selected according to experiences from previous experiments to achieve a potent stimulation of proliferation without or with only minor suppression of adipogenic differentiation [15, 16, 18, 37].

On day 1, after removal of the culture medium, the serum remnants were diluted and the remaining erythrocytes removed by washing the cells in 0.9% NaCl. The culture flasks from every donor and generation were cultured under the following four culture conditions: A) DMEM, 10% human SCP, 100 U/ml penicillin-0.1 mg/ml streptomycin; B) DMEM, 10% human SPPP, 100 U/ml penicillin-0.1 mg/ml streptomycin; C) DMEM, 10% human SPPP, 1 nM bFGF, 100 U/ml penicillin-0.1 mg/ml streptomycin; and D) DMEM, 10% human SPPP, 1 nM EGF, 100 U/ml penicillin-0.1 mg/ml streptomycin.

The cells were maintained at 37°C in an atmosphere of 5% CO2, and the medium was changed every 2–3 days. After reaching confluence, the cells were trypsinized, the specific serum-containing culture medium (A, B, C, or D) was added to stop trypsin activity, and the cell suspension was centrifuged at 700g and room temperature for 7 minutes. The pellet was resuspended in its specific culture medium, and the cells were seeded into wells of 2 cm2: 12 wells, each containing 5,000 cells per cm2 for the assessment of proliferation activity, and four wells, each containing 30,000 cells per cm2 for the assessment of differentiation capacity. For a stem culture, 10,000 cells per cm2 were seeded in a 25-cm2 tissue culture flask.

The medium was changed every 2–3 days. When the cells in the stem culture reached confluence, they were seeded again according to the above-mentioned scheme. This procedure was continued until generation 5.

Stimulation of Differentiation of Preadipocytes into Adipocytes

Preadipocytes were kept in M199, 10% human serum (SCP in medium A, SPPP in B, SPPP + 1 nM bFGF in C, and SPPP + 1 nM EGF in D), 100 U/ml penicillin-0.1 mg/ml streptomycin until they had reached confluence. Then, to induce differentiation, the medium was changed to DMEM, 5% human serum (SCP in medium A, SPPP in B, SPPP + 1 nM bFGF in C, and SPPP + 1 nM EGF in D), 1 μM insulin (Roche Diagnostics, Manheim, Germany,, 0.1 μM cortisol (Sigma-Aldrich), 0.5 mM IBMX (Sigma-Al-drich), and 100 U/ml penicillin-0.1 mg/ml streptomycin for 7 days. The medium was changed three times over this period. On day 8, the medium was changed to DMEM, 5% human serum (SCP in medium A, SPPP in B, SPPP + 1 nM bFGF in C, and SPPP + 1 nM EGF in D), and 1 μM insulin. The medium was changed three times up to day 14.

Cell Culture and Seeding

To assess proliferation activity, samples of 10 patients were tested, and for the assessment of GPDH activity, six of these samples were analyzed. Each sample was cultured under the four different above-mentioned culture conditions (A, B, C, and D) for five generations. For each sample, generation and culture condition proliferation activity was assessed by calculating doubling time and differentiation capacity by measuring GPDH-specific activity [34, 43].

Because cells cultured with SCP showed severe problems in proliferation activity in passages 1 and 2 already, differentiation was assessed in these first two generations only via analysis of GPDH activity. All data shown are based on repetitions with six or more samples.

Assessment of Proliferative Activity

Cells were trypsinized every other day from day 1 to day 11 after cell seeding. Cells of two wells were pooled, cells were counted three times in Neubauer cell chambers, and the mean number of cells per cm2 was calculated. The results were plotted against the preceding culture time, and a growth curve was generated. At the logarithmic phase of growth, a regression line was plotted, and the time in hours required to reach 15,000 and 30,000 cells was calculated. The doubling time was determined by subtracting these times.

Assessment of Adipogenic Differentiation

Cell Morphology and Accumulation of Fat Droplets.

Cell morphology was visibly analyzed every other day during proliferation by light microscopy regarding cell shape and cytoplasmatic structure (e.g., expression of stress fibers and adhesion to the culture surface). Furthermore, cells were assessed for visible accumulation of fat droplets throughout the adipogenic differentiation process.

Measurement of GPDH Activity.

Adipogenesis was quantified biochemically via GPDH activity, as previously described by Wise and Green [34]. The concentration of cell protein was measured according to the method of Lowry et al. [44]. Specific GPDH activity was determined by dividing the GPDH activity of each sample by its protein concentration.

Anti-Perilipin Staining.

Adipogenic differentiation was also evaluated by immunofluorescence staining with an anti-perilipin antibody (Progen Biotechnik GmbH, Heidelberg, Germany, Perilipin is a phosphoprotein localized exclusively on the surface of intracellular lipid or cholesterol storage droplets in mature adipocytes and steroidogenic cells. It cannot be found in undifferentiated preadipocytes [4547].

The preadipocytes were plated onto sterile silanized 76 × 26-mm slides (Engelbrecht Medizin & Labortechnik GmbH, Eder-muende, Germany, at an average density of 20,000 cells per cm2. The cells were maintained at 37°C in an atmosphere of 5% CO2, and the medium was changed every 2–3 days. After cells had grown to confluence, the slides were fixed using 100% acetone for 20 minutes at 4°C and air-dried. Antibodies were diluted with antibody diluent (DAKO, Glostrup, Denmark, The primary antibody was incubated for 1 hour at room temperature. After careful rinsing with phosphate-buffered saline (PBS; Bio-chrom), the secondary antibody was applied and incubated for 1 hour at 37°C in complete darkness. After careful rinsing with PBS, the cell nuclei were stained with 4′,6-diamidin-2′-phenylindol-dihydrochloride (DAPI; Roche Diagnostics) at room temperature for 15 minutes. The slides were coated with glass cover slips (Menzel-Gläser, Braunschweig, Germany, by using Mowiol (Calbiochem, San Diego,

Statistical Analysis

As interindividual differences in differentiation capacity and proliferation activity of human preadipocytes are known [35, 40], samples of 10 different patients of both sexes and a broad range of age were analyzed. Student's t test for paired samples was applied to the results obtained.

PDGF-Enzyme-Linked Immunosorbent Assay

To determine the amount of PDGF-AB in the SCP and SPPP samples, a PDGF AB-enzyme-linked immunosorbent assay (ELISA) was performed (R&D Systems Inc., Minneapolis, with each serum sample in triplicate.


Cell Morphology

Cells cultured with SCP tended to have a round cell shape instead of the spindle cell-like shape of those preadipocytes cultured with SPPP. Cell adhesion to the culture surfaces was less than in cultures with SPPP. Furthermore, they regularly expressed intraplasmatic stress fibers. With stimulation of differentiation into adipocytes, preadipocytes cultured with SCP visibly accumulated fat droplets only in a very limited amount. In contrast, cells cultured with SPPP increasingly accumulated fat droplets according to the ongoing differentiation process, especially when bFGF or EGF was supplemented.

Proliferation Activity (Fig. 1)


Cell number doubled after a mean of 2,250.6 hours (SD ± 1,498.4) in generation 1 and 984.9 hours (SD ± 418.2) in generation 2.


Doubling time varied between a mean of 199.81 hours (SD ± 122.3) in generation 3 and 428.4 hours (SD ± 448.1) in generation 4.


With SPPP supplemented by 1 nM bFGF, doubling time varied between 18.48 hours (SD ± 10.5 hours) in generation 1 and 162.9 hours (SD ± 129.9) in generation 5.


Adding 1 nM EGF to SPPP stimulated proliferation with a doubling time of 24.6 hours (SD ± 11.6) in generation 1 and 218.6 hours (SD ± 350.7 hours) in generation 5.

Statistical Analysis of Proliferation Activity with Student's t Test for Paired Samples

The use of SPPP, with or without addition of EGF or bFGF (media B, C, and D), showed significantly higher proliferation activity than the culture containing SCP (medium A). Culture conditions C and D showed significantly faster cell proliferation than culture condition B in generations 2, 3, and 5 (C) and 1, 2, 3, and 5 (D), respectively (p ≤ .05). There was no statistical difference between culture conditions C and D at any time.

Differentiation Capacity

Accumulation of Fat Droplets (Fig. 2).

Cells cultured with SPPP, with or without supplementation of bFGF or EGF, visibly accumulated fat droplets in accordance to their expression of GPDH. Cells cultured with SCP only accumulated fat droplets very rarely.

GPDH Activity (Fig. 3).

In SCP cultures, cell confluence great enough to stimulate differentiation was only achievable in generation 1. Specific GPDH activity in the stimulated wells was 13.7 mU/mg (SD ± 7.9), whereas it was 34.3 mU/mg (SD ± 26.3) in the nonstimulated wells. In SPPP cultures, the mean specific GDPH activity was in the range 13.6 mU/mg (SD ± 3.5) to 577.5 mU/mg (SD ± 530.1) in the stimulated wells, and it was 28–12.3 mU/mg in the nonstimulated wells. In SPPP + bFGF cultures, the mean specific GPDH activity was a maximum of 924.6 U/mg (SD ± 1451.6 mU/mg) in generation 1 and a minimum of 15.94 mU/mg (SD ± 16.63 mU/mg) in generation 5. In the nonstimulated wells, the specific GDPH activity detected ranged from 12.4 mU/mg in generation 3 to 5.9 mU/mg in generation 5. In SPPP + EGF cultures, the mean specific GPDH activity ranged between a maximum of 487.5 mU/mg (SD ± 384 mU/mg) in generation 1 and a minimum of 29 mU/mg (SD ± 24.7 mU/mg) in generation 5. In the nonstimulated wells, values ranging from 5.87 (SD ± 4 mU/mg) to 3.7 mU/mg (SD ± 3.5 mU/mg) were recorded.

Statistical Analysis of GPDH Activity with Student's t Test for Paired Samples

There was no statistically significant difference between culture conditions B, C, and D (p ≤ .05). Due to low proliferation activity, there was insufficient data based on culture condition A to allow a statistical comparison between SCP and SPPP with or without bFGF or EGF.

Perilipin Expression (Fig. 4).

Adipogenic differentiation was also evaluated by immunofluorescence staining with anti-perilipin antibodies. Adipocytes differentiated from preadipocytes cultured with SPPP (with or without bFGF or EGF) regularly showed a broad expression of perilipin throughout the expanded cell monolayers; cells cultured in SCP only expressed perilipin in a very limited amount (data not shown).


Performing a PDGF-ELISA, the average concentration of PDGF-AB in SCP (16.7 ± 7.3 ng/ml) and SPPP (1.2 ± 0.6 ng/ml) could be determined. These results match those of other groups [1, 2, 11, 48].


As far as we know, the use of human SPPP for the culture of primary human preadipocytes has not been previously documented. Establishing a culture system on the basis of human SPPP is a key tool in the enhancement of adipose soft tissue engineering. Because SPPP can easily be obtained via a conventional preoperative blood donation [14], it will allow a completely autologous system in the future. Platelet-poor plasma is a special form of plasma, produced from whole blood by separating plasma from red blood cells using a high spin to pellet platelets with the red cells. It is almost completely free of platelets and PDGF [13]. As a potent inhibitor of adipose differentiation and the most important stimulator of proliferation in human serum, PDGF is secreted by aggregating platelets during the process of coagulation [49]. When platelet-poor plasma coagulates, almost no PDGF is released because of the low platelet count. Therefore SPPP has a low PDGF level. In our study, we could determine the PDGF level of SPPP at 1.2 (± 0.6) ng/ml, which was much lower than that of SCP (16.7 ± 7.3 ng/ml).

Vogel et al. have shown that PDGF is the growth-limiting molecule in 3T3 cell cultures that have reached confluence with serum from complete blood [49]. Only PDGF or serum from complete blood was able to significantly increase proliferation activity in these cell cultures, whereas the addition of SPPP only resulted in a slight increase in DNA synthesis. The authors attributed this to the low PDGF level of SPPP [49]. In contrast to that, the data obtained in our experiments revealed that cultures with conventional serum had a significantly lower proliferative activity than those with SPPP, despite their higher PDGF levels. This might be due to different ways of preparing SPPP with subsequent differences in levels of growth factors in the two experiments. Furthermore, there are known differences in proliferative characteristics and the reaction to growth factors between 3T3-L1 cells and primary cells, which might also account for these observations [17, 35, 40, 42].

PDGF, however, not only stimulates the proliferation of primary human preadipocytes but also inhibits their differentiation into adipocytes [15, 16, 49]. By using SPPP, it should be possible to avoid the inhibiting effects of PDGF on adipose differentiation. Our results determined a markedly higher GPDH activity in those cells cultured with PDGF-poor SPPP than in those cultured with SCP and higher PDGF levels, thus confirming these results.

Reichert et al. could determine that the stimulating effect of fetal calf serum on preadipocyte proliferation was inversely correlated with the capacity to undergo terminal differentiation. This inhibitory effect of serum on differentiation seemed to be significantly higher in human preadipocytes than in those of other species [41]. Entenmann and Hauner found that in the absence of bovine serum, only 4% of primary human preadipocytes proliferated [42]. Exposure to serum produced a significant increase in cell number, whereas under serum-free conditions, the differentiation activity of cells was greater than under serum-containing culture conditions. These results show that serum contains components that stimulate proliferation and inhibit differentiation. In subsequent experiments, Entenmann and Hauner demonstrated that 0.1–50 ng/ml PDGF induced a significant, dose-dependent reduction in GPDH activity in preadipocytes in serum-free cultures [42]. These results again pointed towards PDGF being an important inhibitor of adipogenic differentiation in serum.

Exposure to bFGF (0.1–100 ng/ml) in the same serum-free system was also followed by a significant, dose-dependent suppression in GDPH activity, but one without any effect on cell number or morphology, thus conflicting with the results of the current study [15]. In our study, the addition of 16.5 ng/ml bFGF (1 nM) to a medium containing SPPP significantly stimulated preadipocyte proliferation, and a considerable increase in GDPH activity compared with SCP or SPPP without added growth factors (up to a maximum of 924.55 mU/mg in generation 1) was observable. Our results are supported by the findings of Hutley et al., who have shown that exposure of primary human preadipocytes to FGF-1 and FGF-2 at a concentration of 1 ng/ml in the presence of 10% fetal calf serum results in a significant increase in both proliferative activity and adipogenic differentiation capacity [17]. In 3T3-L1 cells, proliferation and differentiation could not be increased under similar culture conditions [17]. These adverse results might be explained by the fact that Entenmann et al. were working in a serum-free system, whereas Hutley at al. used, as we did, a serum-containing medium [17, 42]. Either bFGF-induced proliferation and differentiation might be supported by further unknown factors in serum, or Hauner's system contained factors inhibiting the known proliferative effect of bFGF.

In our study, SPPP supplemented with 6 ng/ml EGF (1 nM) raised GPDH activity 36-fold compared with SCP and stimulated cell proliferation significantly compared with SCP or SPPP without supplementary growth factors. These results differ from those obtained by Hauner et al., who described a dose-dependent, significant reduction in GPDH activity when preadipocytes in serum-free culture were exposed to 0.1–100 ng/ml EGF during differentiation [15]. The cells retained their spindle cell-like morphology, and cell number increased enormously. This difference in respect to stimulation of differentiation could be due to the fact that Hauner et al. were using a different culture system [15]. After an initial short period of contact with 10% fetal calf serum (first 16–20 hours), Hauner et al. worked with a serum-free system [15]. The differentiation process was induced differently based on a serum-free basal medium with less insulin, supplementation of triiodothyronine, and less IBMX for only 3 days. Differentiation was analyzed after 16 days. Our system worked with 5% human serum, higher IBMX levels for 7 days, and no added substances other than insulin after 1 week. Differentiation was assessed after 14 days. It is likely that serum also contains factors other than PDGF, which either stimulate adipose differentiation or reduce a possible inhibitory effect of EGF on differentiation.

Serrero tested the impact of EGF and FGF on primary rat preadipocytes in a serum-free, differentiation-stimulating culture system [18]. EGF stimulated proliferation but inhibited differentiation into adipocytes, observable through low, dose-dependent GDPH levels. In quantities up to 1 nM, bFGF stimulated proliferation and differentiation, thus again supporting our findings [18]. Regarding proliferation and differentiation characteristics, this may point to bFGF being a better substitute for the missing proliferative activity of PDGF in serum-free or PDGF-depleted culture conditions. However, human preadipocytes may not necessarily react in the same way as rat adipocyte precursor cells, and there are known differences between their respective reaction to various growth factors [2]. Therefore, our results need to be interpreted with care.

The low number of results of differentiation analysis using SCP allow no discussion on a statistical level. Nevertheless, all samples cultured with SPPP show a much greater GPDH activity than those cultured with SCP.


SPPP represents an ideal component for the culture of primary human preadipocytes. In comparison to SCP, it significantly raises both the proliferation activity and the ability to differentiate, especially when bFGF or EGF is added. At the same time, establishing a culture system on the basis of SPPP provides a key tool in achieving a fully autologous system in adipose tissue engineering, as SPPP can be easily obtained from a conventional preoperative blood donation.

Figure Figure 1..

Proliferation activity under four different culture conditions over five generations, measured by cell doubling time. Abbreviations: bFGF, basic fibroblast growth factor; EGF, epidermal growth factor; SCP, serum from conventional human plasma; SPPP, serum from platelet-poor plasma.

Figure Figure 2..

Adipogenic differentiation in preadipocytes cultured under different conditions. A+, B+, C+, D+, cells cultured with medium A, B, C, or D after 14 days of stimulation for differentiation (adipocytes with accumulated fat droplets and rounded cell shape). A−, B−, C−, D−, nonstimulated control (preadipocytes cultured with medium A, B, C, or D without stimulation of differentiation; spindle-shape-like cells without visible intracytoplasmic fat droplets).

Figure Figure 3..

Differentiation capacity under four different culture conditions over five generations, quantified by specific GPDH measurement. Abbreviations: bFGF, basic fibroblast growth factor; EGF, epidermal growth factor; GPDH, glycerol-3-phosphate dehydrogenase; SCP, serum from conventional human plasma; SPPP, serum from platelet-poor plasma.

Figure Figure 4..

Anti-perilipin staining. Adipocytes differentiated from pre-adipocytes cultured with SPPP stained using an anti-perilipin antibody, nuclei stained with 4′,6-diamidin-2′-phenylindol-dihydrochloride (×200). Number and size of intracellular lipid storage droplets as indicator for adipogenic differentiation can be determined by green fluorescein isothiocyanate fluorescence.


The authors indicate no potential conflicts of interest.


We thank Tanja Wollersheim and Dr. rer. nat. Harald Staiger for excellent technical advice and support. This work was funded by the European Union via the project Adiporegeneration, contract G5RD-CT1999-00111, project GRD1-1999-11159.