Abstract
Introduction
Mesenchymal stem cells (MSC) based therapies from different sources (e.g. bone marrow, adipose tissue) are currently under intense investigation in pre-clinical and clinical trials and have proven to be a valuable tool in the field of regenerative medicine [10, 42]. These cells have been reported to originate from a perivascular source [13, 14] and have been identified in various tissues (e.g. bone marrow, adipose tissue, skin, muscle, dental pulp) [22]. Adipose tissue represents an easy accessible source, from which MSCs can be released by digestion with a collagenase and isolated by centrifugation [45]. The application of a selected MSC subpopulation requires a second therapeutically intervention but enables preservation for later application and also in vitro tissue engineering purposes. However, cell culturing bears the risk of contamination and induces significant changes in the expression of surface proteins and the cell phenotype [30]. Direct application of the heterogeneous stromal vascular fraction (SVF) harvested from adipose tissue has become an alternative for autologous cell-based therapies since immediate application does not require cell culture procedures. Minimal criteria for the phenotypical identification of SVF cells is the antigen pattern CD34+/CD31-/CD45-/CD235a- which was proposed by the International Society for Cellular Therapies (ISCT) [7].
One example for a currently practiced SVF-therapy is the cell assisted lipotransfer (CAL) for soft tissue augmentation as currently tested for aesthetic or reconstructive indications. The supplement of SVF cells to the fat autograft have reported with an increased graft survival and vascularization accompanied with fewer complication rates [28]. Re-injection of lipo-aspirate results in an unpredictable loss of graft volume between 20% to 90% due to a lack of vascularization, central necrosis and fibrosis [8]. The beneficial effect of applied SVF is predominantly attributed to paracrine mechanisms supporting angiogenesis and anti-apoptosis without immunological concerns since patient own cells are solely applied [20]. However, controversy results of the CAL technique exist reporting increased engraftment [44] but also no significant difference when compared to unmixed lipo-aspirate [35].
This might be contributed to the fact that the studies used different extraction techniques and application protocols making it difficult to directly compare studies in this area [3]. In particular, age, donor site, harvesting technique [15, 39] and extraction process [2] have been identified to have an impact on the yield and viability of the SVF-cells. The adequate proportion of SVF-cells and lipo-aspirate also have an impact on the graft survival and integration, however the optimum concentration of re-injected cells is still under discussion [40]. Hence, for general clinical application further research data have to be obtained to create standardized protocols and safety guidelines.
The UNiStationTM (Neogenesis Co. Ltd, Seoul, Korea) as a novel commercially available semi-automated device for the extraction of SVF and CAL might enable further research and clinical application with a reasonable price. The system consists of a heatable centrifuge with a shaking function (Fig. 1) and a sterile processing compartment created by syringes and transfer pieces (Fig. 2). In the present study we aimed to investigate the quantity and quality of the SVF obtained with the device and compare the results with an established manual laboratory method (“hand preparation”).
Material and methods
Tissue harvesting
Six patients undergoing elective liposuction gave written consent for harvesting subcutaneous fat tissue, which was in accordance with the guidelines of the Declaration of Helsinki for biomedical research and approved by the institutional ethics committee of the University Medical Center of Regensburg (Nr. 08/117) according to the guidelines of the Journal for Clinical Haemorheology and Microcirculation [17]. Lipo-aspirate was obtained by tumescent infiltration followed by liposuction. Thereafter, the lipo-aspirate from each patient was divided in two equal portions for the extraction with the novel medical device and the common laboratory method as control.
UNiStation TM
The device was first preheated to 37°C and 40 ml of lipo-aspirate was processed in the UNiSyringesTM by centrifugation at 700 g for 5 minutes. The tumescent solution and blood cells concentrated in the lower part of the syringe were discarded. 5 mg of the MNP-S Liberase were dissolved in a syringe containing DMEM with a volume equal to the discarded tumescent solution. The dissolved Liberase was added to the remaining tissue in the UNiSyringeTM via the transfer piece (Fig. 2c). The shaking plate was applied to the centrifuge and the tissue was digested at 37°C for 30 minutes with shaking at 2 g. Immediately after incubation, the shaking plate was removed and the syringe was centrifuged with 800 g for 5 min. The SVF aggregated at the bottom of the UNiSyringeTM and the syringe was uncapped with prior upward pulling the plunge by 1 mm to prevent cell loss in the syringe cap [33]. 5 ml containing the SVF were transferred to a new syringe and 35 ml of PBS were added and finally centrifuged at 800 g for 3 minutes. After centrifugation, the lower 5 ml containing the SVF, was transferred to a sterile 50 ml centrifugation tube. The MNP-S Liberase activity was blocked by 5 ml culture medium (αMEM containing 20% FBS). The tube was centrifuged at 300 g for 5 minutes and the resulting SVF pellet was used for further testing.
Preparation with common laboratory equipment (“hand preparation”)
Lipo-aspirate was transferred into a sterile centrifugation tube, 1 ml of DMEM was mixed with 0.125 mg MNP-S Liberase and added in equal volume to each ml of lipo-aspirate. The tissue mixture was incubated on a shaker with 100 rpm at 37°C for 45 min. The digested tissue was vigorously pipetted up and down for ten times in a 25 ml serological pipette to release the cells. The suspension was filtered through a 100μm sterile filter system and then centrifuged at 500 g for 5 min. The supernatant containing tumescent, debris and blood with an oil layer on top was carefully removed with a serological pipette. The remaining pellet containing the SVF was re-suspended in 5 ml of PBS and 5 ml of culture medium (αMEM containing 20% FBS), to stop the collagenase activity. Centrifugation and cell pellet washing with 10 ml PBS were performed and supernatant was removed resulting in a SVF pellet for further investigation.
Flow cytometry analysis
SVF cell pellets were re-suspended in 15 ml of erythrocyte lysis buffer prior to labeling, containing 168 mM Ammoniumchloride (NH2Cl), 10 mM Potassium-hydrogencarbonate (KHCO3), 1 mM Ethylenediaminetetraacetic acid (EDTA lysed at pH 8.0) and 0.5μg/mL 4′,6-diamidino-2-phenylindole (DAPI). The suspension was incubated at room temperature for 15 min followed by centrifugation for 10 min at 300 g. The supernatant containing the lysed erythrocytes was discarded.
Each erythrocyte free pellet was split into four 2 ml Eppendorf cups and resuspended in 100μl PBS. Antibodies against the surface markers CD34, CD45 (both BD Pharmingen, Heidelberg, Germany) and CD271 (Miltenyi Biotec, Bergisch Gladbach, Germany) conjugated to different fluorescent dyes were added to one cup. The appropriate matching isotype control was added to the remaining cup whereas all steps were performed in the dark. Following incubation for 30 minutes at room temperature, 1.5 ml of PBS were added to each cup and mixed gently by pipetting up and down. Thereafter, each cup was centrifuged for 5 minutes at 300 g and the supernatant with non-bound antibodies was discarded. The cell pellet was re-suspended in 700μl of PBS and 100μl of the AccuCheck counting bead solution (InvitrogenTM) whereas the isotype control cell pellet samples were re-suspended in 800μl PBS.
Each sample was vortexed for 10 seconds to ensure the homogeneity of its suspension prior to measuring by flow cytometer (BD FACSCanto II, BD Biosciences). At least 30.000 events of the three isotype control samples and at least 250.000 events of the test sample were detected prior to event recording. The pressure was adjusted in order to not exceed 3000 events per second to ensure precise measurement.
Vital MSCs were gated due to physical appearance in the FSC-SSC dot plot. The cell number of positive labeled cells for each marker was investigated by counting beads. The composition of the SVF was investigated for the percentage of single positive and double positive cells. All gates were set according to the matching isotype control.
CFU-assays
ASCs derived from SVFs extracted with both methods were seeded in passage 2 with defined densities of 5, 10, 15, 20, 25 and 30 cells/cm2 onto polystyrene culture dishes with 100 mm diameter. The cells were incubated with culture medium at 37°C under an atmosphere containing 5% CO2 for 14 days. The culture medium (α-MEM) was supplemented with 20% Fetal Bovine Serum (FBS), 1% GlutaMAX and 100 U/ml of Penicillin and 0.1 mg/ml Streptomycin and medium was changed every third day [37]. On day 15, the cells were fixed for 5 min with a 4% Paraformaldehyde (PFA) solution and stained for 30 min with a 0.05% crystal violet solution (CV). Excess stain solution was removed by washing the culture dishes twice with distilled water. Finally, the dishes were air-dried in an inverted position at room temperature under non-sterile conditions. All visible colonies were counted and marked with a pen to prevent miscounts of colonies.
Statistics
The sample pairs were tested using the paired t-test. Unpaired data was compared with unpaired t-test. Differences between all samples processed by the hand preparation (h) or using the novel medical device (d) were tested by the one-way analysis of variance (one way ANOVA). P-Values <0.05 were considered statistically significant, p-Values 0.05 < p < 0.08 were considered as tendency. Linear correlations are described with the Pearson’s r.
Results
SVF cell quantity
A tendency towards a lower cell concentration per gram of lipo-aspirate was apparent for the medical device regarding CD34 (d: 5.7×104±6.0×104 vs. h: 13×104±11×104; p = 0.06). No significant differences between both methods were found for the cell concentration per gram of lipo-aspirate, regarding CD45+ (d: 3.1×104±3.6×104 vs. h: 3.3×104±2.5×104; p = 0.82) and CD271+ (d: 1.7×104±2.2×104 vs. h: 3.4×104±4.0×104; p = 0.11) (Fig. 3). An estimation of the overall cell concentration per initial gram of lipo-aspirate shows a tendency towards a lower outcome with the medical device (d: 1.1×105±1.1×105 vs. h: 2.0×105±1.7×105; p = 0.06).
Composition of the SVF
A significant lower percentage of CD34+ cells was found for the novel medical device compared to the control group were cells were extracted by hand preparation. However, the CD34+ cells harvested by the device had a higher tendency for being positive for CD271+ when compared to the hand preparation method. The percentage of CD45+ positive cells was higher but without a significant difference. No double positive cells for CD45+/CD271+ were found regardless of the extraction method (Table 1).
CFU-assays
Higher seeding concentrations reached significantly higher amounts of colonies per area (d: p < 0.001, h: p = 0.003). However, the CFU capacity (colonies per initial seeded cells) was independent of the seeding concentration for both methods (d: p = 0.50, h: p = 0.93), showing a linear Spearman’s correlation of r = – 0.341 for preparation with a medical device and r = – 0.287 for the hand preparation. The overall CFU capacity was 12.45% ±3.70% for the cells extracted with the use of the medical device and similar for hand prepared cells with 10.4% ±4.71% (p = 0.147).
Discussion
Application of stem cells or SVF requires reproducibility of the cell extraction method not only for safety reasons but also to provide data that can be compared with clinical outcome among different studies. All preparations performed with the medical device and the common laboratory method provided cell concentrations in the typical range from 1*105 to 5*105 cells per gram lipo-aspirate as previously described [2, 43]. An optimal cell concentration of the SVF for CAL application has not been established yet, but increased graft survival was reported for using yields between 105 and 107 cells per ml re-injected lipo-aspirate [40]. Yoshimura et al. suggested for best graft retention that half of the harvested lipo-aspirate should serve for SVF extraction and subsequently used with the remaining lipo-aspirate for reinjection [44]. This recommendation is integrated in the novel medical device since 800ml of lipo-aspirate can be centrifuged at the same time and subsequently a maximum of 400ml can be processed for the extraction of SVF.
The beneficial effects of the SVF are not only based on their transdifferentiation potential but also linked to their paracrine secretion which inhibits inflammatory response at first and subsequently promotes tissue repair [6, 11]. In addition, a higher secretion of cytokines and chemokines was found for the whole SVF than in selected subpopulations of the SVF [26]. This suggests that the absolute amount of SVF cells is less important than the overall amount of paracrine factors by the SVF cells.
The quality of the SVF obtained with the medical device showed similar results as reported for freshly isolated SVF in previous studies, regarding the percentage of CD34+ as well as CD45+ and CD271+ positive cells [23, 38]. A lower percentage of CD34+ cells was found the SVF when using the medical device for preparations. Nevertheless, the preparation with the medical device provides a considerable percentage greater than 50% for 34+ cells which is comparable with reported range for freshly isolated SVF [23, 38]. This is of importance given that CD34 is used to identify hematopoietic and endothelial stem cells [34] that are promoting vasculogenesis which is mandatory for the process of regeneration [4, 32].
CD34+ positive cells co-expressing the marker CD271 have been described as a highly proliferative subpopulation with a vast differentiation potential [9] that were also found in a considerable percentage in this work [36]. Cells selected only by the marker CD271 decrease immunoreactions by paracrine mechanisms and enhance cellular engraftment [27]. Donor site and age influence the amount of CD271+ cells in the freshly isolated SVF although they can always be obtained in a considerable amount [15].
The tendency towards a higher percentage of CD45+ cells can be most likely attributed to the relative lower percentage of CD34+ cells with the medical device, as the CD45+ cell concentrations do not differ between both methods. Although, the leukocyte common antigen (CD45) was used to identify leukocytes on the one hand and on the other hand as a negative control for adipose tissue-derived stem cells [7], a pro angiogenetic effect and a higher formation of neo-vessels have been reported for SVF containing CD45+ cells when compared to CD45 deprived SVF [32]. This effect might be in part linked to CD34+/CD45+ double positive cells [12] that were also detected in the present study. Noteworthy, the preparation method did not influence the CFU capacity, which was similar to previously reported CFU capacity [7].
We propose that the SVF extracted with the novel medical device could not only be used for the CAL procedure but also for treatments which are currently under investigation for example the treatment of non-unions following bone fractures [19], hypertrophic scars [16] or even the revascularization after a myocardial infarction [29]. However, general application safety regulations have to be established prior to compile treatment protocols. Moreover, the donor site and patients’ age have to be considered as they influence the quality of the SVF and thereby its beneficial potential [1, 15].
Limitations of this study
Some changes of the protocol were made to adapt to the research setting. Releasing the cells from the adipose tissue was acquired by the use of the MNP-S Liberase, which is for research only and not suitable for a clinical setting.
During the flow cytometric analysis no vitality staining was performed and only the physical appearance in the FSC-SSC dot plot was used to determine vital cells.
Conclusion
The UNiStationTM can be a useful tool in the extraction of the SVF and ASCs both in a research and a clinical setting. It offers a SVF comparable in quality to the results by preparation with common laboratory equipment and values known from literature. However, some small improvements of the handling and the tools are needed to ensure constant extraction results.
Besides this fact, the semi-automated process needs little space and is applicable to an operating theater. For a CAL treatment it offers the possibility of processing up to 800 ml of lipo-aspirate at the same time. Most important, due to a closed syringe and transfer system, the novel medical device contributes to ensuring sterility in a clinical setting for future cell therapies in patient care.