Today, clinical cell therapies using ASCs are in progress, and several clinical trials are ongoing  and require more-reliable, reproducible, and safe methods for in vitro expansion of the cells. Therefore, the transition from FBS- or HS-medium supplementation to defined XF/SF culture conditions would be one of the most important steps forward in considering the suitability of ASCs for clinical use. By removing all the animal-derived components as well as the undefined serum from the cell-culture workflow, the safety of the patient receiving cell transplant can be improved.
Traditionally, ASC culture medium has been supplemented with FBS, which is not a preferred option in clinical therapies because of xenogeneic components with critical safety issues [12, 25]. Consequently, different kinds of alternatives for FBS have been studied considering the clinical use of ASCs. Trivedi and colleagues  replaced FBS with 20% human albumin during the ASC expansion for clinical use to treat diabetes, whereas Tzouvelekis and colleagues  used autologous platelet-rich plasma for the cell expansion to treat patients with pulmonary fibrosis. AutoHS is currently used for the expansion of ASC by our group for the reconstruction of bone defects in the craniomaxillofacial area [4, 28]. However, as mentioned earlier, limitations are associated with the use of autoHS or serum derivatives, such as lot-to-lot variability , limited availability, and undefined composition, and therefore, the use of XF/SF medium would be the preferred option.
Studies have been performed on ASCs/BMSCs studying the defined XF- or SF-culture conditions; Dromard and colleagues  demonstrated that ASCs can be expanded as floating spheres in defined SF-culture systems supplemented with 2% human plasma and specific growth factors. Further, Santos et al.  investigated a microcarrier-based bioreactor system for the XF/SF expansion of ASCs and BMSCs. Moreover, the suitability of human platelet lysate (PL) for FBS substitution has been investigated by several groups. Schallmoser and colleagues  introduced a standard protocol for platelet preparation for animal protein-free cultures of ASCs, and Naajikens et al.  showed that PL-cultured ASCs had a similar differentiation capacity and increased proliferation rate when compared with FBS cultures. Blande et al., in contrast, showed that ASC population doubling time in PL cultures was significantly lower than that in FBS cultures, but the immunophenotype was similar, and both cultures retained the differentiation potential of the cells.
Still, a better-defined culture environment is needed, and to our knowledge, this study is the first report describing ASC isolation and expansion in completely XF/SF conditions maintaining the basic stem cell characteristics of ASCs. In the past, XF or SF expansion of mesenchymal stem cells has been reported, but the cell isolation and early expansion and differentiation studies were carried out in serum-containing medium [19, 30, 33–35]. In this study, we isolated the cells without serum exposure in XF/SF conditions by using BD PureCoat carboxyl flasks. Onward from passage 1, the cells were able to grow on basic Nunclon Δ surface vessels in XF/SF conditions when either CELLstart CTS coating or XF/SF CM was used. Considering the future clinical applications, it is an advance that the cells are isolated and expanded in fully XF/SF culture conditions instead of using serum supplementation at any point of the culture. If the patient is exposed to undefined components under in vitro expansion, an increased risk occurs for cross-contaminations and immune reactions in a patient receiving the cell transplant. Nevertheless, patient safety is still the most important aspect considering clinical use of ASCs.
In addition to safety, it is advantageous for in vitro expansion of cells to be performed in a shorter time scale. Our studies on the cell-proliferation rate were consistent with the results of earlier studies of mesenchymal stem cell cultures under XF or SF conditions [19, 33, 34] in which the higher proliferation rate of XF- or SF-cultured cells compared with FBS cultures was demonstrated. In our study, the cumulative population doubling in XF/SF medium was superior when compared with both FBS and HS conditions. Efficient expansion of ASCs in XF/SF conditions is crucial for clinical sustainability where a large cell number is required in a minimum time scale.
Cell surface-marker expression profile of ASCs was largely similar between cells grown under different culture conditions, except the significant difference in the expression of CD54 (ICAM-1), which is a marker of endothelial cells and cells of the immune system. To our knowledge, CD54 expression of ASCs has not been studied earlier in XF/SF conditions. The lower expression of CD54 may suggest that a more homogeneous cell population is achieved through more-selective isolation and expansion protocols compared with cells isolated in the presence of serum. In addition, weaker cell adhesion under XF/SF conditions was observed, which may reflect on lower expression of the adhesion molecule ICAM1 (CD54). This aspect of XF/SF cultures and the possible selective effect on cell population has to be investigated in subsequent XF/SF studies, especially when a coating is used.
Some variations were also seen in the expression of CD11a (integrand am), CD14 (lip polysaccharide receptor), CD19 (leukotriene B4 receptor), and CD86 (costimulatory molecule for T-cell activation) on cells grown in XF/SF conditions versus serum-containing medium. All of these markers are known to interact with immune-related cells, and therefore, the culture conditions may affect the immunogenicity of ASCs.
The minimal criterion for the immunophenotype of MSCs described by Dominici et al.  was defined for cells cultured under standard condition in a medium with FBS supplementation. However, our current results with XF/SF cells demonstrate that the cell surface-marker profile applies to ASCs cultured under XF/SF conditions as well. The hematopoietic progenitor cell marker CD34 was moderately expressed in both XF/SF and serum-supplemented conditions in contrast to earlier described criteria. However, similar results for CD34 expression have been reported by others [36, 37], and the variable interpretations could be explained by differences cell-culturing and -passaging protocols. In the current study, the expression of CD34 and CD54 was higher in passage 2, but the expression level was decreased in later passages, indicating a more homogeneous population.
The chondrogenic differentiation capacity of ASCs cultured under XF/SF conditions was strong compared with cells cultured in serum-containing medium based on the Alcian blue staining. Similar findings have been shown by Chase et al. , in which a robust chondrogenesis in SF-culture conditions was seen when compared with serum-containing medium. In our study, the Alcian blue staining of proteoglycans was intense in cells expanded under XF/SF conditions, and also the pellet size after micromass-culture was larger compared with serum-containing medium, which was in agreement with Chase’s results. This result promises potential use of ASCs in chondrogenic applications, but further research is needed to investigate the chondrogenic-differentiation potential of ASCs in XF/SF conditions.
Moreover, the differentiation potential toward osteogenic and adipogenic cells was investigated in XF/SF conditions. Unlike in previous studies [28, 31, 32], osteogenesis and adipogenesis was induced in totally XF/SF differentiation media, and serum was substituted by XF/SF supplement of the STEMPRO MSC SFM kit. When ASCs were cultured under XF/SF conditions, they showed moderate differentiation potential toward osteogenic and adipogenic cells, as demonstrated by ALP and Oil Red-O staining, as the differentiation was not as efficient as seen in serum-containing medium. By optimizing the differentiation protocols for each condition, the efficiency of osteogenic and adipogenic differentiation can be enhanced. Furthermore, the reduced differentiation potential may be due to decreased cell adhesion for cells undergoing differentiation under XF/SF conditions. The weak cell-attachment hypothesis was supported by altered morphology and cell detachment during the proliferation studies, as well as decreased enzymatic digestion time. Another explanation is that ASCs need a more nutrient-rich media to promote robust differentiation. The follow-up differentiation studies with FBS- or HS-based media strongly indicate that nutrients play a key role in efficient differentiation. These findings are consistent with previous studies in which the importance of cell adhesion and nutrients during the cell differentiation has been shown [38, 39].
In addition, the responses seem to be donor specific, and possibly, some cell lines respond better to the osteogenic induction, whereas others respond better to adipogenic induction. Thus, variation in the responses exists, and different stages of differentiation are evident, as shown by the high standard deviations of gene expressions. Also, the mRNA expression and enzymatic activity may not be in line because the regulation occur on posttranscriptional and translational levels, and finally, on the level of formation of an active enzyme. The cells cultured in different conditions may also be at different stages of their differentiation process, thus not expressing the same markers simultaneously. Still, the activity increase in gene-expression level shows commitment to osteogenic or adipogenic pathway, depending on donor cell line and culture condition.
Nevertheless, although at an early stage, differentiation occurred in XF/SF medium, showing that the cells have the capacity for trilineage differentiation, as shown by Oil Red-O, ALP, and Alcian blue staining. Chase et al. demonstrated that BMSCs expanded in SF conditions and differentiated in serum containing induction medium retained their ability to differentiate into adipocytes, chondrocytes, and osteoblasts. In addition, Yang et al. published similar results on the differentiation potential of ASCs expanded in a hypoxic XF environment. The cells expanded in XF medium had equal multilineage differentiation capacity, as compared with cells expanded in traditional serum-containing medium when serum induction was used during the differentiation. Taking these previous results into account, the efficiency of differentiation could easily be improved by serum induction during the culture in differentiation medium.
However, the aim of this study was to remove all the undefined components from the cell-culture workflow, and therefore, serum induction is not a preferred option for use during differentiation. Conversely, autoHS supplementation may be used in clinical treatments for differentiation induction, and the cells would still be expanded in defined XF/SF conditions before differentiation. Also, ASCs can be implanted to the defect site in their undifferentiated state, and the cell differentiation then occurs in vivo, as has been performed by our group [4, 28].
Additionally, growth factors and biomaterials can be used to support cell differentiation in XF/SF conditions. Cordonnier and co-workers  showed that growth-factor induction is especially effective for cells cultured in low serum. In their study, the effect of bone morphogenetic proteins (BMPs) -2, -4, and −7 on osteogenic differentiation of BMSCs was evaluated in low (2%) and higher (10%) FBS-supplemented cultures, leading to a conclusion that BMP-4 induction in low-serum cultured cells was the most effective osteogenic inducer. Similarly, the osteogenic differentiation potential of XF/SF cultured ASCs could be enhanced by growth-factor induction. Furthermore, the differentiation capacity of XF/SF-cultured ASCs can be enhanced by inductive biomaterials, such as osteopromoting bioactive glass  or β-tricalcium phosphate , or alternatively, chondrogenesis-inductive materials such as 3-D woven polycaprolactone scaffolds .