Keywords

7.1 Introduction

A stem cell is defined by its ability to self-renew and to differentiate along multiple lineage pathways. Stem cells have the potential to develop into different cell types in the body during their lifespan. Even after long periods of quiescence, stem cells retain the ability to divide, and the two daughter cells have the potential to either remain a stem cell within the stem cell niche or to develop into a more specialized cell with specific functions. Stem cells can broadly be classified into two categories, namely adult stem cells and pluripotent stem cells. Pluripotent stem cells include embryonic and induced pluripotent stem cells and will not be discussed further in this chapter. Adult stem cells are further subdivided into two categories, namely, haematopoietic stem cells (HSCs) and mesenchymal stromal cells (MSCs). The latter contains a small population of stem cells. MSCs are present in most tissues in the body and are required to restore normal tissue function via repair and regeneration mechanisms (Jones et al. 2002). The bone marrow, Wharton’s jelly present in the umbilical cord and adipose tissue are the most common sources used to isolate MSCs.

The clinical translation of MSC research remains a major challenge. Research groups around the world are working on various strategies to apply MSCs in a variety of settings. There is currently a need for consensus in standardization regarding isolation and expansion procedures, characterization of the cells, evaluation of multipotency, including tissue lineage induction, the best route of administration and monitoring of engraftment success.

Stem cells should meet various criteria before being considered for regenerative medicine purposes. The cells should (1) be present in abundant quantities in the tissue of interest, (2) require minimally invasive harvesting procedures, (3) differentiate along multiple cell lineage pathways that are reproducible and can be regulated/controlled, (4) be safe and effective when transplanted in an autologous or allogeneic manner and (5) be manufactured according to GMP guidelines (Gimble et al. 2007).

In 2013, the International Society for Cellular Therapy (ISCT) reviewed the nomenclature and characterization guidelines for adipose-derived stromal cells (ASCs) and in a joint statement with the International Society for Cellular Therapy (ISCT ) deemed it important to distinguish between the adipose-derived stromal vascular fraction (AD-SVF) and culture expanded ASCs (Bourin et al. 2013). SVF refers to the cellular component present in adipose tissue after mature adipocytes and peripheral blood cells have been removed during the isolation procedure. AD-SVF consists of a heterogeneous cell population which includes preadipocytes, MSCs, endothelial cells, leukocytes, adipose tissue-associated macrophages and other cell types. ASCs refer to the adherent multipotent cell population within the AD-SVF. It remains to be established whether AD-SVF cells or culture expanded ASCs are better suited for clinical applications.

This chapter will discuss AD-SVF isolation procedures, ASC expansion conditions, phenotypic characterization and qualitative and quantitative techniques used to assess the differentiation capacities of ASCs into the adipogenic, osteogenic, chondrogenic and myogenic lineages.

7.2 Isolation Methods

MSCs have been successfully isolated from nearly all postnatal organs and tissues. To date, bone marrow-derived MSCs (BM-MSCs) are the best described, particularly in the clinical setting. These cells serve as the gold standard to which other sources of MSCs are compared (Kern et al. 2006). Adipose tissue, like the bone marrow, derives from the embryonic mesodermal layer (Zuk et al. 2001). Interestingly, adipose tissue is one of the tissues in the human body with the highest number of stem cells, containing between 100- and 1000-fold more multipotent cells per volume unit when compared to the bone marrow. It has been shown that about 3.5 × 104 preadipocytes can be isolated from 1 g of adipose tissue (Fournier and Otteni 1983; Ersek and Salisbury 1995; von Heimburg et al. 2004; Strem et al. 2005). Fraser and colleagues (2006) also demonstrated that adipose tissue yields a 500-fold greater number of ASCs when compared to the MSC yield obtained from the bone marrow. Liposuction procedures can yield anywhere from 100 ml to >3 l of lipoaspirate, which is routinely discarded as biological waste. A recent review mentioned that 20 mg of lipoaspirate should yield >1 × 107 SVF cells (Banyard et al. 2015). This translates to an average expected yield of 1 × 106 ASCs, as it is estimated that 15–30 % of the AD-SVF consists of ASCs (Bourin et al. 2013). The ease with which these cells can be harvested and abundant cell yields render adipose tissue a preferred stromal/stem cell source (Fraser et al. 2006).

The isolation of ASCs from adipose tissue was first pioneered in the 1960s. Minced rat fat pads were extensively washed to remove contaminating HSCs , then incubated with collagenase and centrifuged to obtain an AD-SVF pellet. The selection for plastic-adherent fibroblast-like cells from the AD-SVF concluded this isolation process (Rodbell 1966; Rodbell and Jones 1966). ASCs, at that time referred to as resident MSCs in human adipose tissue, were first described by Zuk and colleagues in 2001 (Zuk et al. 2001). The initial procedure of mincing human adipose tissue by hand was simplified by the development of liposuction surgery.

7.2.1 Manual Isolation Procedure

In 2006, Dominici and colleagues (Dominici et al. 2006) suggested that the standard isolation protocol developed by Zuk and colleagues (2001, 2002) should be accepted as an established methodology to obtain the AD-SVF from lipoaspirate. In 2010, Estes and colleagues published a detailed method, adapted from the method described by Zuk and colleagues (Zuk et al. 2001), to isolate ASCs (Estes et al. 2010). This commonly accepted approach involves five basic steps, namely, (1) the removal of contaminating peripheral blood through washing steps, (2) enzymatic digestion of the adipose tissue, (3) separation of SVF from mature adipocytes, (4) lysis of red blood cells present in SVF and (5) selective isolation of the adherent cell component present in AD-SVF. Isolation procedures may differ concerning the enzymatic digestion step, type and composition of culture medium, approaches used to lyse red blood cells and cell seeding density. Data comparisons between research groups become problematic due to these different approaches. Other factors that may influence the quality of a cell therapy product include the donor’s age, the location (subcutaneous vs. visceral) of the adipose tissue and the surgical procedures used for tissue harvesting (Gimble et al. 2007; Mizuno 2009; Fossett et al. 2012). These factors should be considered when developing an isolation protocol in order to obtain a reliable source of cells that are safe, free from contamination and are of a high quality for application in the clinic. Please refer to the supplementary material for a brief description of the manual isolation procedure.

7.2.1.1 Seeding

Seeding refers to the cells being placed in a culture flask to allow proliferation (expansion). Seeding density is the number of cells seeded onto a culture surface to ensure a standardized concentration of cells per culture and is reported as the number of cells per cm2. The initial seeding density of the AD-SVF is higher than the subsequent seeding densities that will be used during the expansion phase. This ensures that a sufficient number (15–30 % of AD-SVF) of stromal cells are introduced into the culture flasks. Approximately 5 × 105 cells/cm2 of the AD-SVF is usually used and decreased to 5 × 103 cells/cm2 during the ASC expansion phase. Accurate cell quantification is important to ensure that the correct number of cells have been seeded in the culture flask.

Cell counts may be obtained by using either a manual method such as the trypan blue (0.4 %) dye exclusion method or automated cell counting devices, including benchtop flow cytometers. Benchtop flow cytometers are usually not volumetric and provide an absolute count (cells/μl) when a known concentration of counting beads is added to the sample. The Accuri benchtop flow cytometer (BD Biosciences, San Jose, USA) is an exception; it is a volumetric system that allows for direct determination of an abso lute count without the addition of counting beads.

Various flow cytometry counting beads are available commercially, including Flow Count™ counting beads (Beckman Coulter, Miami, USA) and CountBright™ absolute counting beads (Invitrogen/Molecular Probes, Life Technologies, Eugene, USA). Counting beads are commonly referred to as fluorospheres with a broad excitation/emission range. Certain flow cytometers, such as the FC500 and Navios flow cytometers (Beckman Coulter, Miami, USA), have algorithms built into the instrument software that will perform the absolute count calculations in the background and report the absolute cell counts as the number of cells/μl. It is crucial in these flow cytometric systems that the volume of beads to the volume of cell suspension is the same (1:1; vol/vol) to ensure the reporting of an accurate absolute count. The absolute count can also be calculated manually by using Eqs. (7.1) or (7.2) below.

Equation (7.1) is used to manually calculate the absolute count when the ratio of beads to cells is 1:1 (vol/vol):

$$ \mathrm{Absolute}\ \mathrm{cell}\ \mathrm{count}\ \left(\mathrm{cells}/\mu \mathrm{l}\right)=\left[\begin{array}{l}\left(\frac{\mathrm{number}\ \mathrm{of}\ \mathrm{events}\ \mathrm{in}\ \mathrm{area}\ \mathrm{of}\ \mathrm{in}\mathrm{terest}}{\mathrm{number}\ \mathrm{of}\ \mathrm{bead}\ \mathrm{events}}\right)\\ {}\kern0.75em \times \mathrm{calibration}\ \mathrm{factor}\ \left(\mathrm{known}\ \mathrm{bead}\ \mathrm{concentration}\right)\end{array}\right] $$
(7.1)

Equation (7.2) is used in situations where the ratio of sam ple volume to bead suspension volume is not equal [Eq. (7.2)]:

$$ \mathrm{Absolute}\ \mathrm{cell}\ \mathrm{count}\ \left(\mathrm{cells}/\mu \mathrm{l}\right)=\left[\begin{array}{l}\left(\frac{\mathrm{number}\ \mathrm{o}\mathrm{f}\ \mathrm{events}\ \mathrm{in}\ \mathrm{area}\ \mathrm{o}\mathrm{f}\ \mathrm{in}\mathrm{t}\mathrm{erest}}{\mathrm{number}\ \mathrm{o}\mathrm{f}\ \mathrm{bead}\ \mathrm{events}}\right)\\ {}\times \left(\frac{\mathrm{bead}\ \mathrm{concentration}\ \mathrm{assigned}\ \mathrm{t}\mathrm{o}\ \mathrm{specific}\ \mathrm{l}\mathrm{o}\mathrm{t}}{\mathrm{volume}\ \mathrm{o}\mathrm{f}\ \mathrm{sample}\ \mathrm{in}\ \mathrm{u}\mathrm{l}}\right)\end{array}\right] $$
(7.2)

Please refer to the supplementary material for a brief descri ption of the trypan blue (0.4 %) dye exclusion assay as well as an example of how an absolute count is obtained using a benchtop flow cytometer.

7.2.1.2 Expansion of ASCs In Vitro

Several investigators suggest that seeding density may affect cell proliferation. There is once again no consensus among investigators regarding the seeding densities used, although most investigators use a seeding density of 5 × 103 cells/cm2 during the ASC expansion phase. The effect of seeding density on MSC proliferation was demonstrated with BM-MSCs that were seeded at 1 × 102 and 5 × 103 cells/cm2 (Both et al. 2007). The cells seeded at the lower density reached the target of 2 × 108 cells in total in a shorter time period compared to cells seeded at 5 × 103 cells/cm2. The cells seeded at the higher density took approximately 4 days longer to achieve the same target. Similar results were observed by Lode and colleagues in 2008 using synovial fat pad-derived MSCs seeded on three-dimensional scaffolds (Lode et al. 2008) and by Fossett and colleagues (2012) who showed that low seeding densities increase the proliferation capacity in vitro.

7.2.1.2.1 Cell Expansion for Research Purposes

The in vitro expansion of primary cells such as ASCs to reach therapeutic numbers requires that growth conditions mimic the physiological environment in vivo. Such an environment is achieved in vitro through the use of a synthetic basal culture medium enriched with growth factors, hormones and other necessary nutrients (Freshney 2006; Freshney 2010). In addition, ASC proliferation may be stimulated by several exogenous supplements including fibroblast growth factor 2 (FGF-2), sphingosylphosphorylcholine, platelet-derived growth factors and others (Song et al. 2005; Kang et al. 2005; Chiou et al. 2006; Jeon et al. 2006; Mizuno 2009).

Traditionally, ASCs are expanded and maintained in culture medium (DMEM or α-MEM) containing serum. The use of serum during ASC expansion plays a vital role in ASC attachment, longevity and proliferation as it contains essential components such as amino acids, growth factors, hormones, lipids, vitamins, adhesion factors, binding proteins, spreading factors and other trace elements (Lennon et al. 1995; Lennon et al. 1996; Van Der Valk et al. 2004, 2010). Foetal bovine serum (FBS) is widely accepted as the stand ard serum supplement for in vitro studies in the research and experimental setting (Sotiropoulou et al. 2006; Chen et al. 2009). Many disadvantages such as batch-to-batch variation and xeno-immunization are however associated with the use of FBS. Another associated risk is the transmission of zoonotic disease through contamination with Mycoplasma, viruses and prions (Lennon et al. 1995, 1996; Van Der Valk et al. 2004; Kocaoemer et al. 2007; Van Der Valk et al. 2010; Chieregato et al. 2011; Kølle et al. 2013; Kyllonen et al. 2013).

Preparation of cellular therapy products under GMP conditions requires the minimal use of animal-derived products in the manufacturing process (Müller et al. 2006; Crespo-Diaz et al. 2011). Consequently, several human and chemical alternatives are being explored in GMP manufacturing of ASC cell therapy products. Human-derived alternatives such as human serum (Stute et al. 2004), platelet lysate (Schallmoser et al. 2007) and platelet-rich plasma (Doucet et al. 2005) (Table 7.1) may potentially result in an enhanced culture environment that more accurately mimics the in vivo environment (Azouna et al. 2012). However, their relative efficacy is still largely debated, and there is currently no consensus on which of the alternatives is the best to replace FBS (Baer et al. 2010; Bieback et al. 2012; Patrikoski et al. 2013; Koellensperger et al. 2014).

Table 7.1 A comparison of the most commonly used human alternatives to FBS
Full size table
7.2.1.2.2 Xeno- and Serum-Free Cell Expansion for Clinical Application

Synthetic, chemically defined serum-free medium has been recommended as a potential replacement to animal-derived ser um as it contains the necessary growth factors required to sustain ASC expansion (Van Der Valk et al. 2004, 2010). Nonetheless, many studies have revealed that when using serum-free medium, ASCs are incapable of adhering to plastic without the addition of coating agents. In addition, ready-to-use serum-free medium is expensive and might therefore be a less cost-effective alternative for the replacement of FBS in the clinical setting (Lennon et al. 1995; Lund et al. 2009; Baer et al. 2010; Rajala et al. 2010; Yang et al. 2012; Kyllonen et al. 2013; Patrikoski et al. 2013).

7.2.2 Isolation Using Automated Closed Systems

An automated closed system allows for minimal handling of a sample by the operator, resulting in a potentially safer, less labour-intensive isolation procedure. Contamination is a highly undesirable event in the clinical setting and is limited through strict control of standard operating procedures (SOPs) and processing according to GMP (Gimble et al. 2011). Although the manual isolation procedure is relatively easy to reproduce and is carried out in a sterile environment, there is a higher contamination risk when compared to a closed system. The introduction of contaminating particles into the culture system is significantly reduced using closed systems. Hicok and colleagues suggest that in order to comply with the regulations involved in a clinical trial, an automated system for cell isolation should be used (Hicok and Hedrick 2011). Commercial companies such as Cytori Therapeutics Inc (San Diego, USA) and Biosafe Ltd (Lake Geneva, Switzerland) have designed state-of-the art cell-processing devices that allow for the isolation of the AD-SVF from lipoaspirate in automated closed systems. It has been suggested that automated systems may contribute to the advancement of clinical research and the translation thereof into clinical trials.

The Sepax® cell-processing system (Biosafe Ltd, Eysins, Switzerland) was developed to isolate and concentrate nucleated cells from the umbilical cord blood, peripheral blood or bone marrow (Zingsem et al. 2003; Aktas et al. 2008; Zinno et al. 2011). The successful application of the system in the HSC field resulted in the system also being validated for the isolation of AD-SVF cells from human adipose tissue (Guven et al. 2012). The Sepax® system automates the cell washing and concentration steps during adipose tissue processing. The device is comprised of the main Sepax® sample processing unit, single-use sterile disposable sample processing kits and ready-to-use application-specific software protocols.

In Asia, Australia and Europe, but not in the USA, the Cytori’s Celution® sample processing system has been granted regulatory approval for the use in adipose tissue processing and AD-SVF cell isolation. Multiple clinical trials are ongoing in several countries using the Celution® system (Akita et al. 2012; Houtgraaf et al. 2012). This automated system was specifically designed for fat processing and to mimic the modified manual isolation steps from the method de scribed by Zuk and colleagues (Zuk et al. 2001, 2002). The Celution® system is a fully closed automated system, which uses a single-use sterile disposable kit comprised of a processing canister, connection tubes and waste bags. The Celase® enzyme is used for adipose tissue digestion instead of collagenase. One of the advantages of the system is that it performs several quality control processes before the automated sample processing is initiated. For example, upon insertion of the sample processing kit into the device, the Celution® system performs a wet test to validate the integrity of the kit by checking for any leakages in the system and consequently ensures that the sample will be processed in a fully closed manner. After the lipoaspirate suspension is introduced into the processing canister, the system automatically weighs the sample and performs rinsing steps with Ringer’s lactate solution to remove blood cells and other contaminants. The Celution® system automatically calculates the amount of Celase® enzyme required for proper digestion of the volume of adipose tissue processed. The instrument will indicate the volume of Celase® enzyme which needs to be added manually. Enzymatic digestion takes place automatically through constant agitation. Following digestion and separation, the fraction of cells released from the adipose tissue matrix is pumped into the centrifuge chamber where it is washed and concentrated into the output chambers on either end of the centrifuge chamber. The final AD-SVF cell product is then aspirated from the output chambers (~5 ml).

Evaluation of the different automated systems needs to take into account factors such as the ability to process a large variety of lipoaspirate volumes and the degree of automation it provides. Practical considerations include the cost of the device and the consumables required. Sample processing time (which is influenced by the volume of lipoaspirate introduced) may also be a consideration. It should be noted that the Sepax® system is not fully automated as adipose tissue digestion needs to be carried out manually. However, an automated adipose tissue protocol has been developed for the isolation of the AD-SVF following digestion. In comparison, the Celution® system i s a fully automated system validated for the processing of adipose tissue and requires minimal intervention by the operator.

7.3 In Vitro Characterization of ASCs

In 2006, the ISCT released a position statement defining the minimal criteria required to identify ASCs (Dominici et al. 2006). According to these guidelines, ASCs and BM-MSCs share the same characteristics, namely, that they are multipotent, plastic-adherent cells that express specific surface antigens and have the capacity to differentiate into adipocytes, chondrocytes and osteoblasts. These criteria became the gold standard for all in vitro studies involving ASCs. However, in 2013 a new set of criteria was suggested by the IFATS and the ISCT. The overall criteria remained essentially unchanged, except that the report suggested a repertoire of protein surface markers that will allow investigators to distinguish between AD-SVF, ASCs and BM-MSCs (Bourin et al. 2013).

7.3.1 Immunophenotypic Characterization of ASCs

The challenge in phenotyping ASCs is that none of the surface expression markers used is specific to ASCs (Table 7.2). Therefore, to be more confident in the phenotypic assessment of ASCs, it is advisable to follow a multicolour flow cytometric approach, where multiple surface protein markers are simultaneously stained with various fluorochrome-conjugated antibodies to provide a more accurate co-expression profile of the cells (Zimmerlin et al. 2013; Bourin et al. 2013; Baer 2014; Donnenberg et al. 2015). According to the IFATS criteria, more than 80 % of the cells should co-express the following markers: CD13, CD29, CD36, CD44, CD73, CD90 and CD105 (Bourin et al. 2013). There should be less than 2 % haematopoietic cells as well as endothelial cells present in an ASC isolate. Consequently less than 2 % of the cells should stain positive for CD31 (endothelial cells), CD45 (leukocytes) and CD235a (erythrocytes) (Bourin et al. 2013). Simultaneous staining of cell surface markers allows for better discrimination of ASCs from other cells that may be present in the isolation, but do not adhere to the above-mentioned co-expression profile.

Table 7.2 A summary of marker expr ession profiles and known physiological functions of the more common surface antigen markers used to phenotype adipose-derived stromal cellsa
Full size table

7.3.2 In Vitro Differentiation Assessments and Limitations

It is well established that ASCs or subpopulations thereof have the ability to differentiate into adipocytes, osteoblasts, chondrocytes and myocytes. The differentiation capacity of ASCs suggests a role in the turnover and maintenance of adult mesenchymal tissues (Caplan 2009). In vitro differentiation of ASCs into these lineages (adipocytes, chondrocytes, osteoblasts and myocytes) is therefore traditionally used to confirm the multipotent characteristics of the isolated ASCs. However, true stem cell plasticity entails the ability of these cells to form tissues from different germ layers. Investigators have shown that under appropriate culture conditions, ASC subpopulations display stem cell plasticity by differentiating into non-mesodermal tissues such as neurons (Ning et al. 2006) and epithelial cells (Brzoska et al. 2005). However, a significant amount of research is still needed to confirm the purported plasticity of ASCs. Effective clinical application of ASCs will rely on understanding the biological effectors that are responsible for maintaining a specific differentiation state. In addition to the criteria set out by the IFATS and ISCT above, it has been suggested that histochemical staining techniques used to establish differentiation are qualitative and that a quantitative approach, such as quantitative reverse transcription polymerase chain reaction (RT-qPCR), is required. The minimum number of genes required to confirm the differentiation of ASCs into adipocytes, osteoblasts, chondrocytes and myocytes, respectively, is summarized in Table 7.3.

Table 7.3 Minimum gene expressi on requirements to confirm in vitro adipogenic, osteogenic, chondrogenic and myogenic differentiation
Full size table

7.3.2.1 Adipogenesis

High-calorie intake without energy expenditure promotes adipocyte hyperplasia. Signalling factors are responsible for this increase in adipocytes by inducing the conversion of ASCs to preadipocytes and finally differentiation into mature adipocytes (Tang and Lane 2012). ASCs become committed to preadipocytes by bone morphogenetic protein (BMP) -2 and -4 (Huang et al. 2009) and Wnt signalling (Bowers and Lane 2008), causing them to go into growth arrest (Fig. 7.1). Differentiation inducers such as hormones and mitogens (insulin, glucocorticoids and cAMP) then initiate mitotic clonal expansion where the cells’ fibroblastic morphology is lost and the accumulation of cytoplasmic triglycerides occurs, eventually forming mature adipocytes (Tang et al. 2004). Peroxisome proliferator-activated receptor-γ (PPAR-γ) and CCAAT/enhancer-binding proteins (C/EBP) are key regulators of adipogenesis (Kang et al. 2007).

Fig. 7.1
figure 1

Regulation of ASC differ entiation. Fat accumulation, cartilage formation, bone formation and bone resorption are closely related. Adipocytes, osteoblasts and chondrocytes have a common origin, the mesenchymal stem cell, while osteoclasts originate from the haematopoietic stem cell. Osteoblasts regulate bone resorption through the RANKL/RANK/OPG pathway. Osteoblast differentiation is regulated by RUNX2. Key regulators of adipogenesis are PPAR-γ and C/EBP. Adipocytes secrete inflammatory cytokines and hormones which are capable of modulating bone formation. Chondrogenesis is mainly regulated by BMPs, TGF-β and Sox9. Modified from Fig. 7.1 in Cao (2011) and Rosen and Bouxsein (2006). RUNX2 runt-related transcription factor 2, RANK/RANKL receptor activator of nuclear factor-kB ligand, OPG osteoprotegerin, PPAR-γ peroxisome proliferator-activated receptor-γ, C/EBP CCAAT/enhancer-binding proteins, BMP bone morphogenetic protein, TGF transforming growth factor, Sox sex-determining region Y-type high mobility group box (Cao 2011; Rosen and Bouxsein 2006)

Full size image

In the laboratory, in vitro adipogenic differentiation takes 2–3 weeks after initiating differentiation following exposure of the cells to the appropriate induction media . The composition of the adipogenic induction medium is provided in the supplementary material. Adipogenesis is associated with the accumulation of lipid droplets in the cytoplasm of the adipocytes. The number and size of lipid droplets increase as the ASCs differentiate from preadipocytes to mature adipocytes. The lipid droplets are highly organized organelles, consisting of an outer layer of amphipathic lipids such as phospholipids and a core of neutral lipids including triacylglyceride and sterol esters (Guo et al. 2009; Krahmer et al. 2009). Studies using immortalized preadipocytes, such as the mouse 3 T3-L1 cell line, revealed that combinations of 3-isobutyl-methylxanthine with dexamethasone or insulin induce intracellular lipid accumulation. Therefore, to promote the effective differentiation of ASCs into adipocytes, the induction medium is supplemented with IBMX, insulin and dexamethasone. 3-Isobutyl-methylxanthine (IBMX), a cAMP-phosphodiesterase inhibitor, has been shown to inhibit soluble cyclic nucleotide phosphodiesterase activity causing an increase in intracellular cAMP levels by stimulating the cAMP-dependent protein kinase pathway. This increase in cAMP levels is subsequently responsible for the synthesis of lipogenic enzymes during adipogenesis (Spiegelman and Green 1981; Tang and Lane 2012). Ali and colleagues have shown that the removal of IBMX from the induction medium almost completely blocks the normally observed increase in intracellular lipid accumulation (Ali et al. 2006). Insulin acts as a potent adipogenic hormone by triggering the induction of a series of transcription factor, which facilitate differentiation of preadipocytes into mature adipocytes via the insulin-like growth factor 1 (IGF-1) receptor (Ntambi and Young-Cheul 2000). Dexamethasone is a synthetic glucocorticoid agonist used to stimulate the glucocorticoid receptor pathway. Although each agent is able to initiate adipogenesis, an additive effect is observed when the factors are combined (Russell and Ho 1976; Rubin et al. 1978). Please refer to the supplementary material for a brief description of the method used to induce adipogenesis in vitro.

7.3.2.1.1 Assessment of Adipogenic Differentiation

Adipogenic differentiation may be measured using both quantitative and qualitative (qPCR) assays. The minimum number of genes required to confirm the differentiation of ASCs into adipocytes is summarized in Table 7.3. Most of the histological, qualitative assays make use of lipophilic dyes which mainly stain the triacylglycerides present in the lipid droplet core. IFATS and ISCT jointly recommend using the lipid-specific stains Oil Red O and Nile Red (9-diethylamino-5H-benzo[α]phenoxazine-5-one) as histological determinants of adipogenic differentiation (Bourin et al. 2013).

7.3.2.1.1.1 Quantitative Adipogenesis Assessment Using Oil Red O

Oil Red O is a nonfluorescent hydrophobic stain with a high affinity for neutral lipids, such as triacylglycerides, present in lipid droplets (Fig. 7.2). The stain can then be extracted from the cells using isopropanol and quantified by measuring the absorbance values spectrophotometrically, but this method of quantification lacks specificity due to non-specific Oil Red O staining (Ramírez-Zacarías et al. 1992). Please refer to the supplementary material for a brief description of the qualitative method used to evaluate adipogenesis differentiation usi ng Oil Red O in vitro.

Fig. 7.2
figure 2

Adipogenic lineage microscopy analysis. (a) Oil Red O-stained and 1 % Toluidine Blue-counter stained non-induced culture. The red arrow indicates Oil Red O residue overlaying the culture. The cells are fibroblast-like, small, slender and elongated as indicated by the blue arrow and conform to describe ASC morphology. (b) Oil Red O-stained and 1 % Toluidine Blue-counterstained adipogenic induced culture. Visible Oil Red O droplets confirm differentiation into the adipogenic linage. (c) A magnification snapshot of (b). The red arrow indicates the presence Oil Red O residue overlaying the cell. Yellow circles demonstrate adipocytes (cytoplasm completely filled with lipid droplets and also fusion of the droplets with associated lipid vacuole formation). Green rectangle demonstrates a more mature preadipocyte with incomplete lipid droplet formation. Blue triangle demonstrates a preadipocyte with an enlarged triangular shaped cell that conforms to the morphology of adipocytes but does not contain any visual Oil Red O-stained lipid vacuoles

Full size image
7.3.2.1.1.2 Quantitative Adipogenesis Assessment Using Nile Red

Lipid droplet formation during adipocyte differentiation may also be visualized using fluorescent microscopy techniques (Fig. 7.3). Nile Red and Bodipy 493/503 are currently the most common lipid-specific dyes used during fluorescence microscopy assessment of lipid droplet formation (McNeil et al. 1991; Smyth and Wharton 1992; Brasaemle et al. 2000; Lo Surdo et al. 2013; Aldridge et al. 2013).

Fig. 7.3
figure 3

Visualization of adipocyt e formation following adipocyte differentiation using fluorescence microscopy. (a) ASCs were stained with Nile Red (2 μg/ml) and DAPI. (b) ASCs were stained with Bodipy 493/503 (20 μg/ml) and DAPI. Both images were captured at Day 21 after adipocyte differentiation was induced

Full size image
7.3.2.1.1.3 Quantitative Adipogenesis Assessment Using Flow Cytometry

Flow cytometry provides a more quantitative measurement of both the proportion of cells containing intracellular lipid droplets and the degree of lipid accumulation within each cell (Fig. 7.4) (Fink et al. 2004; Schaedlich et al. 2010; Chazenbalk et al. 2011; Aldridge et al. 2013; Ceppo et al. 2014). Most investigators make use of the hydrophobic dye, Nile Red (Fink et al. 2004; Lu et al. 2010; Menssen et al. 2011; Lo Surdo et al. 2013; Aldridge et al. 2013). Nile Red is a solvatochromatic dye, meaning that it can change fluorescent colour in different polar environments (Fowler and Greenspan 1985; Greenspan et al. 1985). Nile Red emits yellow-gold fluorescence (emission >528 nm) when dissolved in neutral lipids such as triglycerides. The fluorescence emission spectrum of Nile Red shifts to the deep-red spectrum (>590 nm) when it dissolves in an amphipathic lipid environment such as phospholipids (Greenspan et al. 1985). However, when dissolved in lipid droplets, the yellow-gold fluorescence is more easily visualized than the deep-red fluorescence. This may be contributed to by the ratio of neutral lipids to phospholipids present in a lipid droplet.

Fig. 7.4
figure 4

Flow cytometric analysis of a dipocyte differentiation of ASCs after staining with Nile Red (20 ng/ml final concentration). Adipocyte differentiation is associated with the intracellular accumulation with lipid droplets, which consequently results in an increase in cellular complexity. Green-coloured events indicate early stages of intracellular lipid accumulation. Orange-coloured events are indicative of cells with higher levels of intracellular neutral lipid content due to the accumulation of intracellular lipid droplets (increase in cellular complexity). Flow cytometric analysis was performed on a Beckman Coulter 3-laser, 10-colour Gallios flow cytometer, after 21 days of incubation of ASCs in adipogenic medium

Full size image

There are also several other hydrophobic fluorescent dyes available that may be useful in the quantitative assessment of adipogenesis, such as Bodipy 493/503. There are reports that Bodipy 493/503 is more specific than Nile Red, but further studies are needed to validate this statement (Gocze and Freeman 1994). A definite advantage of Bodipy 493/503 is that it has a more specific fluorescent emission spectrum range. This allows for easier combination of cell surface marker s with the lipid-specific dyes to study the phenotypic characteristics of adipocytes in more detail.

7.3.2.2 Osteogenesis

There appears to be a complex relationship between adipogenesis and osteogenesis . Both adipocytes and osteoblasts are derived from a common multipotent MSC, thus leading one to believe that the two pathways are interconnected (Fig. 7.1). Obesity increases adipocyte differentiation and fat accumulation and is believed to decrease osteoblast differentiation and bone formation. Several recent investigators have reviewed the complex relationship between adipogenesis and osteogenesis (Cao 2011; Liao 2014; Nuttall et al. 2014; Romagnoli and Brandi 2014; Atashi et al. 2015).

The osteogenic maturation pathway involves cell proliferation, differentiation and matrix deposition followed by mineralization. Differentiation of ASCs into osteoblasts in vitro involves incubating a monolayer of ASCs with ascorbic acid, β-glycerophosphate and dexamethasone for 3 weeks. The composition of the osteogenic induction medium is provided in the supplementary material. Ascorbic acid acts as a cofactor for collagen synthesis through the hydroxylation of proline and lysine residues in collagen and induces extracellular matrix (ECM) production through the synthesis of non-collagenous bone matrix proteins. β-Glycerophosphate is needed for calcification and mineralization of the ECM. Dexamethasone regulates osteoblastic gene expression (Fiorentini et al. 2011). Dexamethasone treatment has been shown to increase alkaline phosphatase activity in vitro which is required for matrix mineralization and morphological transformation to cuboidal-shaped cells (Cheng et al. 1994). Alizarin Red S staining is commonly used for identifying calcific deposition during matrix mineralization in osteogenic differentiation cultures which is an early marker for differentiation (Fig. 7.5). Please refer to the supplementary material for a brief description of the qualitative method used to evaluate osteogenic induction in vitro.

Fig. 7.5
figure 5

Light microscopy analysis of osteo genesis. (a) Alizarin Red S-stained non-induced culture (osteogenesis control) showing no visible calcium depositions. (b) Alizarin Red S-stained osteogenic induced culture showing the red calcium deposits within the bone matrix characteristic of bone formation

Full size image

7.3.2.3 Chondrogenesis

Cartilage is a specialized connective tissue which consists of chondrocytes surrounded by an ECM. Chondrocytes produce and maintain an ECM which is rich in glycosaminoglycans and proteoglycans and interact with collagen and elastic fibres. The cartilage matrix is composed mainly of type II collagen, providing tensile strength, and aggrecan, which provides compressive strength. Cartilage is classified according to its composition. Hyaline cartilage consists mainly of type II collagen fibres, elastic cartilage of type II collagen and elastic fibres and fibrocartilage of type I collagen fibres (Naumann et al. 2002). Chondrogenesis involves the recruitment and condensation of MSCs followed by differentiation into chondroblasts, chondrocyte maturation, and ECM production.

In vitro chondrogenesis is induced by artificially condensing the cells by centrifugation followed by addition of TGF-β3, dexamethasone and BMP-2 to the medium. The composition of the chondrogenic induction medium is provided in the supplementary material. Johnstone and colleagues (1998) developed a culture system to facilitate chondrogenic differentiation for BM-MSCs (Johnstone et al. 1998). This pellet culture system allows for cell-cell interaction similar to those that occur during embryonic development (Fell 1925). It has been shown that a defined medium, to which dexamethasone and TGF-β1 have been added, is required (Johnstone et al. 1998). Dexamethasone induces a metachromatic staining pattern in aggregated cells. This indicates the synthesis of a cartilaginous matrix which was confirmed by collagen type II immunohistochemistry. The use of TGF-β1 either alone or in combination with dexamethasone induces chondrogenesis in aggregated cells. It was later found that the addition of BMP-6 further increased chondrogenesis (Sekiya et al. 2001). Critical roles are played by BMPs in compaction of MSCs and for the shaping of the condensations (Zuscik et al. 2008). Comparisons between BMP-2, -4, and -6 with regard to their ability to enhance in vitro formation of cartilage demonstrated that BMP-2 was the most effective, resulting in increased pellet weight along with more proteoglycan and collagen type II production (Sekiya et al. 2005). However, the addition of BMPs to cartilage induction medium is not necessarily required. Please refer to the supplementary material for a brief description of the method used to induce chondrogenesis in vitro.

More detailed strategies for the assessment of chondrogenic differentiation in vitro include techniques such as light microscopy, transmission electron microscopy and immunohistochemistry (Ichinose et al. 2005). Please refer to the supplementary material for a brief description of the qualitative method used to evaluate the induction of chondrogenesis in vitro. Other studies have shown that after a week in induction medium, cell pellets consist of three layers: the superficial zone, containing fibroblast-like cells; the middle zone, containing apoptotic cells; and the deep zone, containing matrix-producing chondrocyte-like cells. After 14 days, the middle zone disappears, and the deep zone dominates after an induction period of 21 days (Ichinose et al. 2005). Biochemical analysis of chondrogene sis includes methods that quantify the total glycosaminoglycan content in pellets (Naumann et al. 2002; Estes et al. 2010).

7.3.2.4 Myogenesis

The in vivo myogenic differentiation capacity of stem cells is highly sought after in regenerative medicine, as this property brings hope that these cells may restore regenerative capacity in muscular and neuromuscular disorders, myoskeletal tissue trauma, sport injuries and urologic incontinence. Benefits associated with this type of intervention include faster recovery time and a reduced risk of graft rejection, and preclinical evidence supports the application of MSCs across allogeneic barriers (Joo et al. 2014). Several lines of evidence exist for successful in vitro myogenic differentiation of MSCs from various origins (Gang et al. 2004). However, successful clinical translation remains to be convincingly demonstrated.

ASCs display superior myogenic differentiation potential (Stern-Straeter et al. 2014; Zych et al. 2013) and might be an ideal candidate for application in muscle tissue engineering of fibrotic muscle (Choi et al. 2012). ASCs can also be induced to differentiate into cardiomyocytes following induction by 5-azacytidine (Cao et al. 2004; Carvalho et al. 2013). Co-culturing of ASCs with primary myoblasts, in addition to stimulation with dexamethasone and FGF-2, promotes differentiation up to multinucleated myotubes (Beier et al. 2011; Bitto et al. 2013). Although no minimal criteria have been established for the myogenic differentiation potential of ASCs, the molecular markers in Table 7.3 have been used for quantitative assessmen t of myogenic differentiation.

7.3.3 Experimental Animal Models

In order to reach the long-term aim of clinical translation of cell-based therapy, preclinical safety and efficacy need to be demonstrated in a nimal models. Transplantation of in vitro expanded ASCs in the appropriate experimental animal model is therefore an important step in the development of cell therapy products. Sensitive cell-tracking techniques are needed in order to determine the most optimal route of administration as well as the homing ability of the transplanted ASCs. Transduced ASCs containing GFP lentiviral vectors could offer a feasible in vivo tracking system as it has been shown that transduction of ASCs with a GFP-expressing lentiviral vector does not affect their phenotypic expression or their differentiation potential (van Vollenstee et al. 2016).

7.4 Conclusion

Although manual isolation procedures are less costly than closed automated systems, the risk of contamination associated with manual procedures makes them a less attractive option for clinical applications. GMP isolation and expansion procedures require the minimal use of xenogeneic components, and only cell therapy products that are free of animal-derived products will lead to successful translation into the clinic. ASCs are regarded as non-minimally manipulated cells and are classified as a drug by the Food and Drug Administration (FDA). However, it is currently unclear whether in vitro expanded ASCs hold any significant benefit over AD-SVF, particularly in the clinical setting.

One of the major pitfalls in the rapidly expanding field of ASCs is that no inclusive panel of either cellular or molecular markers exists that specifically characterize these cells. The ISCT and IFATS regularly publish updated guidelines for the classification of ASCs. However, the current guidelines are still not specific and are based on the ability of the cells to adhere to plastic, to express a set of non-specific cellular markers and to differentiate into adipogenic, osteogenic, chondrogenic and myogenic lineages.

For example, although classical stains (Oil Red O, Alizarin Red S and Toluidine Blue) confirm lineage differentiation in vitro, it is unclear if all induced ASCs have differentiated into the respective lineage. The translation of in vitro assays to the patient is also unclear, and well-characterized in vivo experimental models are needed to validate the engrafting, homing and differentiation potential of the cell therapy in question. The availability of reliable and reproducible in vivo experimental models will therefore contribute to a more confident and potentially more rapid translation of ASC research to the clinical setting.