The potential use of ASCs in tissue engineering and regenerative medicine has been well demonstrated in multiple pre-clinical animal models [24, 28]. Although ASCs have been characterized based on immunophenotype, cell yield and differentiation properties [26, 29–31], the effects of BMI on ASC functionality have not been fully elucidated. The present study demonstrated that cryopreserved human ASCs from patients with higher BMIs were compromised in colony-forming potential and growth under exposure to low serum concentrations. Overweight BMI-ASCs were also compromised in osteogenic differentiation capacity when compared to lean BMI-ASCs. Cryopreservation was not a factor in the observed differences, as both freshly isolated BMI-ASCs and cryopreserved BMI-ASCs were comparable in both osteogenic and adipogenic differentiation capacities. Moreover, in vitro adipogenesis data suggested that early time point intracytoplasmic lipid uptake was positively correlated to BMI. Exposure of ASCs to low serum uncovered differences among BMI-ASCs that were less apparent at higher serum concentrations. Although one individual donor exhibited variation in the trend of BMI and proliferation (e.g. BMI-ASC 29.4), the determination coefficients were sufficient to form conclusions about the correlation between proliferation and BMI.
Overweight BMI-ASC osteogenesis was reduced independent of serum levels. These results were similar to a published report that the BMSC osteogenic alkaline phosphatase response to mechanical strain was inversely correlated to donor BMI . These findings paralleled clinical data demonstrating that obese patients possessed lower relative bone area and bone mass compared to lean individuals, when adjusted for body weight, , and this may contribute to an increased risk for fractures . Other clinical studies have demonstrated an inverse correlation between body weight and bone density and peak bone mass. In addition, there is evidence of increased risk of implant failure following total joint knee replacement in obese patients [9–11]. An understanding of the impact of fat metabolism on bone precursors and osteogenic capacity will be necessary to advance ASC use in bone repair applications.
The present data suggests that overweight BMI dependent increase in early adipogenesis may be associated with a decrease in osteogenic capacity. Multiple studies suggest that when ASC fate is committed to the adipogenic pathway, osteogenesis is concomitantly down regulated, consistent with observations in BMSCs [25, 32]. This balance is suggested to be partly modulated by the presence and amount of glucocorticoid, the glucocorticoid receptor pathway activity, crosstalk with the Jak/STAT3 pathway, and the subsequent activation of the AP-1 pathway [33, 34]. Other studies have suggested that an inverse relationship exists between donor age and ASC osteogenesis , and that ASCs from male donors exhibit increased osteogenic potential compared to ASCs from female donors in vitro . However, conflicting outcomes were reported in studies comparing donor age and BMSC osteogenic potential in vitro with one study demonstrating an inverse correlation  whereas other studies reported a positive correlation [38, 39].
ASCs from overweight BMI patients (BMI >25) showed an increase in the intracytoplasmic Oil-Red-O staining at early adipogenesis time points, but no significant difference at later time points. Although an increasing trend was observed in early adipogenic potential that correlates to BMI, an individual donor effect may have contributed to the low adipogenesis in donor L070525 (BMI 30.65 kg/m2). Genes that are induced during both early and late stage adipogenesis have been extensively described [20, 21, 33, 40–42]. Yu et. al., 2010  performed time-dependent RT-PCR analyses of genes involved in adipogenesis in 4 donors, BMI 23.1 ± 1.4, age 39 ± 6 years. The adipogenic transcription factor C/EBPα, the adipokines adiponectin and leptin, and lipoprotein lipase (LPL) all displayed a time-dependent increase during adipogenesis, with the greatest percent increase occurring within the initial 3-day period; however BMI had no significant effect on the mRNA expression of the representative genes that were upregulated on day 3 of adipogenesis (PPARγ, Adiponectin, LPL, and aP2). This suggests that BMI has no overall effect on adipogenesis in vitro, and supports our data demonstrating no correlation between BMI and late adipogenesis. However, Schipper et. al.  reported age-dependent differences in ASC adipogenic potential and proliferative capacity in vitro. Further investigation is needed relating the impact of age and BMI on early and late ASC functionality.
It is widely accepted that hyperplasia follows adipocyte hypertrophy in patients during fat mass expansion which requires a gradual replenishment of the pool of adipocytes in obese individuals [44, 45]. One study proposes that BMSCs are recruited and give rise to bone, muscle, and both white and brown adipose tissues in response to the appropriate developmental cues in vivo . However, the ability of ASCs to differentiate into terminally differentiated cells of both white and brown adipose tissues adds further complexity to the interesting dynamic. The formation of newer, small mature adipocytes is thought to occur via differentiation of preadipocytes; however whether the preadipocyte originates from the fat mass, or from the recruitment of the circulating bone-marrow progenitor cells  remains to be determined. In this regard, it should be noted that Koh et. al.  demonstrated that BMSCs in bone marrow transplant mice became resident as phagocytic macrophages in adipose tissues and resembled trans-differentiated adipocytes, but did not express adipocyte markers.
The present study used cryopreserved ASCs that were passaged in vitro, not freshly isolated ASCs. Several studies have investigated the cryopreservation characteristics of ASCs and examined different methods of cryopreservation on ASC biology in vitro [41, 48–50]. Thirumala et al., 2010 [22, 51], showed that the post-thaw viability of ASCs differed significantly based on the method of freezing and rate of thaw, and determined that using an ethanol enclosed jacketed container, with 10% DMSO and 80% FBS in the cryopreservation medium, yielded the highest post-thaw viability. Other studies have examined the effects of various cryoprotective agents on cell viability, proliferation, and osteo- and adipogenic differentiation [47–49]. These studies indicated that proliferation and osteo- and adipogenic differentiation of ASCs could be maintained in vitro post-thaw. Our studies also demonstrated that cryopreservation did not significantly affect ASC adipogenic or osteogenic potential (Additional file 1: Figure S1, panels a-f).
Reports on the proliferation index of freshly plated ASCs versus cryopreserved ASCs are conflicting. James et. al., 2011  compared ASCs freshly harvested from lipoaspirate to cryopreserved ASCs, and measured growth using Trypan blue dye exclusion assay. Their results demonstrated cryopreservation significantly affected in vitro proliferative capacity. However, studies by Deng et. al., 2008 , and Gonda et. al., 2008  suggest that ASC proliferative capacity following cryopreservation is maintained. Calculations of ASC population doubling times in our studies coincide with reports by Deng et. al., 2008, and Gonda et. al., 2008 [52, 53], as there was no significant difference observed between the doubling times of freshly isolated ASCs and cryopreserved ASCs (Additional file 1: Figure S1, panel g).
James et. al., 2011  also compared ASCs freshly harvested from lipoaspirate to cryopreserved ASCs, and measured osteogenic differentiation (alkaline phosphatase staining and qRT-PCR), and adipogenic differentiation (alizarin red staining and qRT-PCR). The results demonstrated cryopreservation significantly affected osteogenic differentiation, both in vitro and in vivo. The use of recombinant proteins such as IGF and BMP, however, were used to mitigate the deleterious effects of the freeze–thaw process on osteogenic potential. It should be noted that BMI was not a tested factor, and no cytokines or signaling factors were added to their differentiation medium. A handful of cytokines and other factors are known to stimulate osteogenic differentiation in ASCs, including vitamin D, which was added in the differentiation medium used in the present study. In addition, the strong determination coefficients derived from performing the nonlinear regression analyses of the proliferation (r = 0.86-0.9) and osteogenic differentiation studies (day 8, r = 0.86; day 16, r = 0.96) suggest donor metabolic status may also have an effect on ASC functionality.
In vivo, near complete wound healing was observed in mice with calvarial defects that were engrafted with fresh human ASCs as compared to groups engrafted with cryopreserved ASCs which exhibited reduced healing . Although freshly isolated ASCs may exhibit beneficial properties for wound healing and other applications, the majority of applications for ASCs will likely use cryopreserved ASCs, therefore it is important to understand the characteristics and limitations of cryopreserved ASCs. Future studies comparing the functionality of freshly isolated ASCs to cryopreserved ASCs within different BMI groupings are needed to determine whether the observed differences exist in freshly isolated cells.
Other possible contributing factors to the outcomes of this study are patient clinical characteristics that were unavailable from patient records, such as donor hormonal status, smoking and other undisclosed patient parameters. Regression analyses were conducted to determine whether a correlation exists between age, passage and the measured endpoints within the study. All R2 values were less than 0.2 and as such reflect no correlation between the known factors and the measured endpoints.