Technique for Obtaining Mesenchymal Stem Cell from Adipose Tissue and Stromal Vascular Fraction Characterization in Long-Term Cryopreservation

Human mesenchymal stem cells are advantageous in both basic and applied research. The use of this adult cell type overpasses ethical issues-compared to the use of embryonic or other cells-being one of the most promising areas of study in autologous tissue regeneration engineering and cell therapy¹, such as the neoplastic area, the treatment of degenerative diseases, and therapeutic applications in the reconstructive surgery area^2,3,4,5. It has been previously reported that there is an abundant source of mesenchymal multipotent and pluripotent stem cells in the stromal vascular cell fraction of adipose tissue^6,7. These ADSC are considered great candidates for use in cell therapy and transplantation/infusion since a considerable number of cells with a strong capacity for expansion ex vivo can be easily obtained with a high yield from a minimal invasive procedure^5,8.

It was also demonstrated that adipose tissue presents a greater capacity to provide mesenchymal stem cells than two other sources (bone marrow and umbilical cord tissue)⁹. Besides being poorly immunogenic and having a high ability to integrate into the host tissue and to interact with the surrounding tissues^4,10, ADSC has a multipotent capacity of differentiation into cell lines, with reports of chondrogenic, osteogenic, and myogenic differentiation under appropriate culture conditions^11,12,13, and into cells, such as pancreatic, hepatocytes, and neurogenic cells^14,15,16.

The scientific community agrees that the mesenchymal stem cells' immunomodulatory effect is a more relevant mechanism of action for cell therapy^17,18,19 than their differentiation property. One of the most significant merits of the ADSC use is the possibility of autologous infusion or grafting, becoming an alternative treatment for several diseases. For regenerative medicine, ADSC have already been used in cases of liver damage, reconstruction of cardiac muscle, regeneration of nervous tissue, improvement of skeletal muscle function, bone regeneration, cancer therapy, and diabetes treatment^20,21.

To this date, there are 263 registered clinical trials for the evaluation of ADSC's potential, listed on the website of the United States National Institutes of Health²². Different protocols to harvest adipose tissue have been established, but there is no consensus in the literature about a standardized method to isolate ADSC for clinical use^23,24. Lipoaspirate processing methods during and after surgery can directly affect cell viability, the final cellular yield²⁵, and the quality of the ADSC population²⁰. Regarding the surgical pre-treatment, it is not well established which surgical pre-treatment technique yields a more significant number of viable cells after isolation or whether the anesthetic solution injected into adipose tissue affects cell yield and its functions²⁶. Similarly, the difference between techniques for obtaining adipose cells can lead to as much as a 70% decrease in the number of viable ADSC²⁰. According to the literature, mechanical treatments to obtain cell populations with high viability-including ultrasound-should be avoided, for they can break down the adipose tissue²⁰. However, the manual fat aspiration method with syringes is less harmful, causing less cell destruction, with tumescent liposuction yielding a significant number of cells with the best quality²⁶.

This technique uses a saline solution with lidocaine and epinephrine that is injected into the liposuction area. For each 3 mL volume of solution injected, 1 mL is aspirated. In this study, the wet liposuction technique was performed, in which for each 1 mL of adrenaline and saline solution injected, 0.2 mL of adipose tissue is aspirated. The use of digestive enzymes, especially collagenase, is common for the process of isolating ADSC.

After the first isolation step in the laboratory, the final pellet is called stromal vascular fraction (SVF). It contains different cell types²⁷, including endothelial precursor cells, endothelial cells, macrophages, smooth muscle cells, lymphocytes, pericytes, pre-adipocytes, and ADSCs, which are capable of adhesion. Once the final isolation is concluded from in vitro cultures, cells that did not adhere to the plastic are eliminated in medium exchanges. After eight weeks of expansion, medium changes, and passages, ADSCs represent most of the cell population in the flasks²⁰. One of the most significant advantages of using isolated adipose-derived stem cells for a possible future therapy is the possibility of cryopreservation. It was demonstrated that cryopreserved lipoaspirate is a potential source of SVF cells even after 6 weeks of freezing²⁸, with biological activity even after 2 years of cryopreservation²⁹, and full capability to grow and differentiate in culture³⁰. However, during the thawing process, a considerable percentage of cells is usually lost³¹. Therefore, the lipoaspirate removal process and the following methods of cell isolation must ensure the highest cell yield.

This study describes a faster methodology for collecting and isolating ADSC, demonstrating high cellular yield and viability for better efficiency of cellular therapeutics. Furthermore, the effect of this improved technique after long-term SVF cryopreservation was evaluated.

The present study is approved by the Ethics Committee of the UNIFESP (protocol number: 0029/2015 CAAE: 40846215.0.0000.5505), performed after obtaining written informed consent from the patients according to the Declaration of Helsinki (2004). The sample of the present study is composed of nine female patients, aged 33-50 years (average age 41.5) and average initial body mass index (BMI) of 24.54 (ranging between 22.32-26.77) (Table 1) who underwent aesthetic abdominoplasty due to excess of skin after pregnancies, at the Division of Plastic Surgery of the Universidade Federal de São Paulo (UNIFESP), Brazil. To reduce bias, the patients were selected as a homogeneous group considering sex, age, and BMI. The datasets used and/or analyzed during this study are available from the corresponding author upon reasonable request.

1. Collection of lipoaspirate

NOTE: This step needs to be performed in the surgery center.

1. Use 4% chlorhexidine gluconate (see Table of Materials) for skin preparation and asepsis.
   1. Perform a 2 mm subcutaneous skin incision (between the sub-dermis and aponeurosis). Insert a Klein cannula of 26 mm 3 G three-hole and 3 mm caliber and a syringe to inject a total volume of 500 mL of an adrenaline solution (1 mg/mL) (see Table of Materials) diluted in saline (1:1,000,000) in the infraumbilical area.
2. Connect a 60 mL syringe to a 26 mm 3 G three-hole and 3 mm caliber liposuction cannula and insert it through the skin incision, locking the plunger to create a vacuum.
   1. Make pushing and pulling movements so that, with the vacuum created, the lipoaspirate remains in the 60 mL syringe.
3. Using a sterile connector with a valve, transfer the 100 mL of the collected lipoaspirate to a 150 mL polyvinyl chloride transfer bag (see Table of Materials).
   1. Pack the transfer bag in a polystyrene box at room temperature (~25 °C) and take it immediately to the laboratory. Do not take longer than 30 min to start the tissue processing.

2. Processing of lipoaspirate

NOTE: This step is to be performed in the laboratory.

1. First, weigh the bag, gauge the temperature with a digital non-contact infrared clinical thermometer, and leave the bag resting for 5 min inside the laminar flow chamber for precipitation of the greasier layers (bubbles) and tissue separation containing the cells of interest.
   1. Perform a series of tissue washes. First wash: inject 100 mL of DPBS with calcium (1x) into the transfer bag and mix it with the hands.
   2. Let it stand for 5 min and remove most of the basal liquid that precipitates.
   3. Discard the basal liquid with a 60 mL syringe attached to the bag adapter. This process must be repeated twice.
2. Add 100 mL of digestion solution to the bag (93 mL of calcium-free DPBS + 60 µL of calcium chloride (1 g/L) + 7 mL of 0.075% sterile collagenase, see Table of Materials) and leave at 37 °C for 30 min under slow stirring.
3. Transfer all the bag's content to four conical tubes of 50 mL and centrifuge them at 400 x g at 22 °C for 10 min.
   1. Remove and discard the supernatant and add 5 mL of Dulbecco's modified Eagle's medium (DMEM) low glucose supplemented with 20% FBS (Fetal bovine serum) to the cell pellet (Figure 1).

3. Counting of the SVF cells

1. Mix a fresh solution of 10 µL of trypan blue at 0.05% in distilled water with 10 µL of cellular suspension for 5 min.
2. Count viable cells in a Neubauer cell counting chamber³² using an inverted light microscope (see Table of Materials) at 20x magnification.
3. Resuspend the cell pellet in a cryoprotective medium (5 mL of FBS + 10% of Dimethyl Sulfoxide - DMSO) at a concentration of 1 x 10⁶ cells/mL.
4. Place 1 mL of this mix in cryovials. Use a freezing container (see Table of Materials) with a cooling rate of (1 °C/min to -80 °C).
   1. Store at -80 °C for 1 year.
   2. After this time, store in standard cassette boxes immersed in the liquid nitrogen vapor phase (-165 °C).

4. Thawing process of the cells

1. Remove the vials from liquid nitrogen and place them immediately in the 37 °C water bath for 1 min.
2. Place the SVF cells in a conical tube with 4 mL of DMEM (low glucose supplemented with 20% FBS) preheated at 37 °C.
3. Centrifuge at 400 x g at 22 °C for 5 min.
4. Remove the supernatant and add 1 mL of DMEM (low glucose) + 10% FBS. Perform immunophenotyping following the steps below.

5. Flow cytometry technique (immunophenotype multiple labeling)

1. Place 1 mL of cell pellet (concentration of 1,000 cells/µL) in five cytometry tubes (200 µL each).
2. Centrifuge at 400 x g at 22 °C for 5 min and discard the supernatant with a pipette.
3. Add 300 µL of Phosphate-Buffered Saline (PBS) (10x), centrifuge at 400 x g at 22 °C and discard the supernatant with a pipette.
4. Prepare five tubes for different marker combinations as follows: 5 µL of CD11B/5 µL of CD19/20 µL of CD45; 5 µL of CD73/20 µL of CD90/5 µL of CD105/20 µL of CD45; 20 µL of CD34/5 µL of HLA-DR/20 µL of CD45; Cell viability assay-5 µL of fluorescent reactive dye. (see Table of Materials) and a tube with unstained cells and PBS as the negative control. Homogenize in a vortex and incubate at 4 °C for 30 min.
   1. Centrifuge at 400 x g at 22 °C for 5 min, discard the supernatant with a pipette, add 500 µL of PBS (10x), and proceed with cell sorting.
      ​NOTE: Five thousand events are acquired per antibody set in the Flow Cytometer of four colors and five parameters and analyzed with CellQuest software.

6. Seeding of passage 1 (P1)

1. Seed 2 x 10⁵ cells in a 75 cm² culture flask.
2. Add 12 mL of DMEM low glucose + 20% of FBS + 10% antibiotic/antimycotic (with 10,000 units penicillin, 10 mg streptomycin, and 25 µg amphotericin B per mL, 0.1 µm).
3. When the cells reach between 80%-90% confluence, perform trypsinization of adherent cells with 2 mL of 0.25% EDTA-trypsin for 3 min.
4. Count cells again (as mentioned in step 3).
5. Perform immunophenotyping again (as mentioned in step 5).

7. Statistical analysis

1. Use Spearman's Rho Calculator³³ to measure the strength of association between the following variables with P < 0.05, as mentioned below.
   1. Select SVF cellular yield and the number of days SVF stays in culture in the first passage (P1) until 80%-90% confluence (days to P1).
   2. Select SVF cellular yield before and after going to P1.
   3. Consider days to P1 and cellular yield before going to P1.
   4. Select SVF cellular yield with the average percentage of confirmed ADSC.
   5. Calculate the percentage of confirmed ADSC and the cellular yield before going to P1.
   6. Determine the BMI and SVF cellular yield.

8. Differentiation assay

1. Perform the differentiation assay following a differentiation kit protocol (see Table of materials). Figure 4 demonstrates the results for Case 1.

The characterization of the nine individuals studied, including their age, weight, height, and BMI, are shown in Table 1.

According to the cellular yield initially presented, the cell volume inoculated in culture was calculated to be as close as possible to the capacity of the 75 cm² culture flask. The sample volume seeded in each case is described in Table 2. Then, according to the initial cellular yield, a variable volume of cells for each sample was determined: 1 mL for samples with higher cellular yield, 1.1 mL for samples with intermediate cellular yield, and 2 mL for samples with lower cellular yield so as to perform more similar cell seeding between cases. When the culture reached about 80%-90% confluence (Figure 2A) (about 7.5 ± 4.5 days), trypsinization of adherent cells was carried out (Table 2 and Figure 2B).

The cellular yield before passage 1 broadly varied even when the same confluence before trypsinization was observed (Table 2). This can be explained by the fact that cells may have grown in layers. Different parameters from the patients' ADSC were also assessed at different periods, as demonstrated in Table 2.

Some samples (Case 1, Case 2, Case 7) could not be evaluated regarding the percentage of confirmed ADSC and the estimated number of ADSC in culture due to bacteria contamination and lack of available cells to perform cryopreserved SVF immunophenotyping. According to the Spearman's Rho Calculator³³, no statistical differences were found between SVF cellular yield and days to P1 (r = 0.37816, p = 0.31561), between SVF cellular yield before and after going to P1 (r = -0.33333, p = 0.38071), and between days to P1 and cellular yield before going to P1 (r = -0.53783, p = 0.13529). Furthermore, no significant differences were observed when correlating the SVF cellular yield with the average percentage of confirmed ADSC (r = -0.02857, p = 0.95716) and between the average percentage of confirmed ADSC and the cellular yield before going to P1 (r = 0.42857, p = 0.3965). Also, the correlation between BMI and SVF cellular yield could not be considered statistically significant (r = -0.46667, p = 0.20539). Table 3 shows flow cytometric data performed on SVF cells cryopreserved. The initial SVF cells contained a subset of positive cells for hematopoietic markers (CD45, CD11b, CD19, HLA-DR)³⁴. From the initial SVF cell population, a particular subgroup expressed CD11b³⁴ and CD19³⁴ stromal cell-associated markers. The levels of CD73³⁴, CD90³⁴, and CD105³⁴ were intermediate between these values. The initial SVF contained a subpopulation of cells positive for stem cell-associated markers (Figure 3). A mean of 79% of SVFs expressed the HSC-associated marker CD34³⁴.

In total, 21 min were necessary for the three washes, 30 min for collagenase digestion, 10 min for centrifugation, and 5 min for cell counting and plating.

Figure 1: Steps from the protocol adipose-derived stem cells isolation. (A) Bag for lipoaspirate transport in a closed system. (B) The step of resting bag repeated three times, after washing. (C) Lipoaspirate after three washes with DPBS. (D) Lipoaspirate after collagenase digestion. (E) Lipoaspirate after digestion is distributed in a 50 mL tube. (F) Digested lipoaspirate after centrifugation. (G) Final process isolation with the pellet with the stromal vascular fraction (SVF). Please click here to view a larger version of this figure.

Figure 2: Morphology and viability of ADSCs. (A) Plastic adherent mesenchymal adipose-derived stem cells at the first passage after isolation at light microscopy. The cells show adhesion to the plastic and fibroblast-like morphology. (B) Trypan blue assay showing viable cells counted in Neubauer chamber using a light microscope. Please click here to view a larger version of this figure.

Figure 3: Subpopulation of cells positive for stem cell-associated markers in SVF of Case 9 after 8 months of cryopreservation. R1: Total cellular region analyzed in FSC (Forward Scatter) x SSC (Side Scatter) (Size x Complexity); R2: CD45 negative region, whose populations CD73, CD90, and CD105 are positive in this region. Please click here to view a larger version of this figure.

Figure 4: Differentiation assay. (A) ADSC differentiation in chondrocytes. (B) ADSC differentiation in osteocytes. (C) ADSC differentiation in adipocytes. Please click here to view a larger version of this figure.

Table 1: Data from the samples of the individuals studied. *BMI: body mass index.

Table 2: Data from different steps of the procedure from the nine patients analyzed. SVF: stromal vascular fraction; ADSC: adipose-derived stem cell; P1: passage 1; na: data not available.

Table 3: Flow cytometry data from six of the patients. (*) From these CD45- cells, % of ADSC with different combinations of stem cell markers was determined. ADSC: adipose-derived stem cell; SVF: stromal vascular fraction; (*) From these CD45- cells, % of ADSC with different combinations of stem cell markers was determined.

Isolation yield
It is well established that the cryopreservation process, frequently required in cellular therapy, results in significant cell loss, sometimes greater than 50%^29,30,35. Thus, a technical improvement for obtaining high initial cellular yield in isolation is fundamental. The collecting method of lipoaspirate and the isolation method of the cells must focus on preserving a greater number of cells, maintaining high viability, and extracting the maximum number of cells from the initial material while accounting for the long-term culture and manipulation of the cells. Therefore, straight culture maintenance is required to keep the cells away from apoptosis, senescence, or genetic instability, since cell therapy is likely to be effective and safe for patients.

To the best of our knowledge, there is no previous article using this set of steps for the isolation of mesenchymal stem cells derived from lipoaspirate, which results in a time-saving and cost-benefit technique. In this study, each methodological step was reasoned according to the literature that showed the highest final cell yield in cellular isolation. The novelty of the technique performed in this study was the use of manual aspiration associated with a series of lipoaspirate washes, with the subsequent bag resting. The collection bag used to transport and process lipoaspirate allowed undigested tissue fragments to not participate in collagenase digestion. The most critical step is adding calcium chloride to the fresh digestion solution as it potentiates the action of collagenase. The time gain is not yet reported in the literature with a cell thawing method that allows viable cells even after long cryopreservation time. The SVF cellular yield found in this study varied broadly from 6.15 to 26.2 x 10⁵ cells/mL with an average of 15.7 x 10⁵ cells/mL. This may be due to the presence of a more significant amount of adrenaline solution suctioned, which may have been higher or lower according to the stage of the surgical procedure and to the number of other known cell types generally found in the SVF. Although some studies have found negative correlations between BMI and ADSC yield^36,37, this study found no significant correlation, like the other two studies^38,39, decreasing the possibility of that being the cause of the incredible variety of SVF cellular yield found in this study. These data show that the lowest SVF cellular yield obtained was 6.15 x 10⁵ cells/mL. Some studies had already measured the efficiency of ADSC isolation according to the surgical technique for obtaining lipoaspirate. One study obtained 0.087 x 10⁵ cells/mL in freshly isolated SVF for the liposuction technique using adrenaline solution (as used in this study) and 0.143 x 10⁵ cells/mL without it²⁶. This work highlighted the significance of the adrenaline solution injection due to the vasoconstrictive effect that decreases intraoperative bleeding and bruising, as the majority of surgeons choose to perform in clinical practice. Another study demonstrated that live ADSCs isolated ranged from 0 to 0.59 x 10⁵ cells/g lipoaspirate harvested, with an average of 0.295 (±0.25) x 10⁵ cells/g tissue³¹. Some studies have tested different ways to achieve higher ADSC yield. One of these studies achieved about 350 x 10⁵ for the method that presented various constituents in collagenase digestion buffer and the use of an orbital shaker⁴⁰. Another study showed 29.7 (±0.2) x 10⁵ cells/mL as the total number of SVF cells from the abdominal area⁴¹. The abdominal area chosen in this study is still the reference area for the best availability and accessibility of lipoaspirate⁴². The body region for lipoaspirate collection, among other factors as donor age and method of the collection chosen, is a strong determinant of the quality of the ADSC yields.

The possibility of microorganisms' contamination causing the unavailability of cells to continue the experiments was the only execution problem that limited the completion of this study. Even using antibiotics and Good Manufacturing Practice requirements, contamination can occur due to the lack of a total aseptic environment to perform the aspiration of fat.

SVF immunophenotyping
According to The Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy³⁴, one of the three minimal criteria to define human mesenchymal stem cells is that the cells must express CD105³⁴, CD73³⁴, and CD90³⁴ and should not express CD45³⁴, CD34³⁴, CD14³⁴, or CD11b³⁴, CD79a³⁴ or CD19³⁴, and HLA-DR³⁴ surface membrane molecules. Mitchell⁴³ tested fresh SVF cells by immunophenotyping and found a maximum of 54% of cells with ADSC surface markers. In this study, the immunophenotyping revealed a higher percentage of confirmed ADSC (non-hematopoietic cells) of up to 78.91% in the SVF after long-term cryopreservation (ranging from 37.95%-78.91% with a mean of 53.68%). Evidence shows that the progenitors of a stem cell population not yet committed do not express the CD34 marker^34,44,45. Depending on the stage of differentiation, the CD34³⁴ negative stem cells can generate not only hematopoietic progenitors but also more specific mesenchymal precursors, such as osteoclasts, chondrocytes, myocytes, adipocytes, and others. Some studies demonstrated the striking plasticity of the primitive stem cell population, composed of cells with stromal cell function and hematopoietic and mesenchymal progenitors⁴⁵. According to the literature, the complete CD34³⁴ functional role in the tissue formation in SVF cells is still unknown⁴⁶. Mitchell⁴³ showed a mean of 60% cells of SVF expressing CD34³⁴ marker, whereas, in this study, the mean was 78.81%. It is known that the expression of CD34³⁴ surface marker decreases along passages.

Adherent cells and differentiation assay
Depending on the number of cells obtained after isolation, the number of cells seeded in the first culture varied. The first culture time for reaching 80%-90% confluence in 75 cm² flasks took an average of 8.4 days and a standard deviation of 7.5 (±4.5) days ranging from 6.6 to 16.1 x 10⁵ cells/mL. It is to be noted that even for the cases with lower cellular yield, the first culture time led to high cellular yield indexes compared to the literature, probably due to the best availability of viable cells maintained during the entire collection and isolation process. One study obtained a cellular yield of 3.75 (±1.42) x 10⁵ ADSC per mL of lipoaspirate within a 4.1 (±0.7) day culture period⁴⁷. Another study demonstrated a yield of nucleated SVF cells of 3.08 (±1.40) x 10⁵ per millimeter with a mean of 6.0 (±2.4) days in the first culture period⁴³. In this study, by estimating the number of ADSC seeded in culture, which varied as a function of the number of cells observed in the SVF from the collection of the same volume of 100 mL of lipoaspirate, it was verified that the number of days to reach P1 had no relation to cell volume. For example, for Case 9, in which 12.2 x 10⁵ cells were incubated, 6 days were required to reach 80%-90% confluence. For Case 6, in which 14.6 x 10⁵ cells were seeded in culture, a more extended period was necessary (10 days) to reach up to the same level of confluence. Perhaps, a minimum ADSC number is enough for the first adhesion period. There may be significant interindividual variation, such as comorbidities, age, and general health status. Some studies in the literature questioned the importance of considering the patients' inter-individual variability for SVF cell yield^48,49.

According to The Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy³⁴, mesenchymal cell must have the ability to differentiate into three different cell types as osteogenic, adipogenic and chondrogenic lineages as it was demonstrated in Figure 4.

Cell viability, cryopreservation, and genetic instability
Different methods have been used to determine viability loss, defined as plasma membrane integrity damage⁵⁰. However, the cultures can also present early apoptotic cells that these approaches can ignore because they maintain an intact plasma membrane, but they are nonviable. The results reported here show a great range in viability marker, from 23.06%-72.34%, with a mean of 47.6% after long-term cryopreservation. Considering that cryopreservation is a significant step concerning cell therapy, recovery of the maximum number of viable and functional stem cells after thawing is one of the priority issues for the success of cell therapy. The literature has shown that at least 50% of cell viability is lost from 1-4 months after cryopreservation⁴³. Notably, the lowest indexes presented in this study are from Case 5, Case 4, and Case 2, which are the oldest cryopreserved samples (about 2 years). Although they presented the lowest viability indexes, they demonstrated high cellular yield in trypan blue dye exclusion assay in fresh SVF. Although the literature supports causes of viability loss, these rates are lower than expected. Temperature fluctuations in cell storage due to technical reasons can cause the increase and accumulation of stress and favor the accumulation of aqueous portions, generating crystals that damage the plasmatic membrane during long-term cryopreservation⁵¹. The literature shows more than 70% viability for samples with the same or longer freezing time. However, the viability check was performed with a different technique than that carried out in this study⁵². Another study showed that longer cryopreservation negatively affects cellular viability³¹, which can be explained by temperature variations in the -80 °C freezer. Hence, cells often need to stay too long in culture, which increases cell cycle stress, bringing risks for genetic instability and consequently compromising cellular therapy. There is also a consensus in the literature indicating some stress factors and how they affect cell cytogenetic stability, which is essential to maintain the prolonged stem cell cultivation required for cell therapy^35,53.

Methodology benefit
To date, literature shows no standardized protocol to isolate ADSC aiming for clinical applications. Most of the studies demonstrate complex, time-consuming protocols²⁴. In this study, the efficiency of the method versus the time demanded to complete the initial cellular yield must be emphasized: about 1.5 h. According to literature, isolation of adipose-derived stem cells can take about 3 h to 8 h^54,55. Thus, the gain of time allied to the high cellular income is critical for regenerative medicine therapeutics advancement. More cell viability assessments should be carried out parallel to those performed in this work to improve this method. Further randomized controlled trials incorporating a more extensive sample using this methodology are required to countersign these results.