Back to Annual Meeting Program

BACKGROUND: Blood vessel bioengineering holds promise for patients without vein suitable for arterial reconstruction or dialysis access but the lack of a ready source of autologous cells from which to construct new vessels remains a limitation. We have engineered vessels for use in a pre-clinical ovine model using smooth muscle cells cultured from native artery wall (aSMC) but an artery biopsy is less practical in the clinical setting. A number of investigators have suggested adipose-derived stem cells (ADSC) provide a less invasive alternative source of SMC but methods and results have varied widely.
METHODS: To determine the extent to which ovine ADSCs differentiate into an aSMC phenotype, we compared cell morphology and growth characteristics, expression of SMC-specific proteins, and the ability of cells to contract type-I collagen gels. ADSC lines were established from the vascular stromal fraction of subcutaneous fat biopsies from three individual sheep and maintained in stem cell media or grown in SMC differentiation media (DM) (α-MEM, 1% embryonic FBS, penn/strep and L-glut) ± TGF-β (10ng/mL). Cells were also exposed to uniaxial stretch during differentiation to determine any additional effects on phenotype (collagen-coated Flexcell plates, 10% strain, 60 Hz).
RESULTS: Each ADSC isolate was first shown to be multi-potent by growing subcultures in osteogenic or adipogenic differentiation media and documenting typical mineralization and lipid accumulation, respectively. ADSC maintained in stem cell media were smaller than control aSMC, grew to confluence one to two days faster, and expressed low levels of the SMC markers α-actin, myosin heavy chain (sm-MHC) and smoothelin. In contrast, after 7 days in DM, ADSC had developed morphology and growth kinetics similar to control aSMC and showed significant up-regulation of all protein markers. TGF-β or cyclical stretch did not further enhance marker expression compared to DM alone. Unlike control aSMC, undifferentiated ADSC did not contract 3-dimensional collagen gels (diam. after 18 hrs: 14.6±0.2 vs. 18.0±0.0 mm, p<0.001). Remarkably, ADSC grown in DM for 7 days contracted collagen gels as vigorously as control aSMCs, while the addition of TGF-β limited contraction slightly (13.0±0.2 vs. 16.8±0.2 mm, respectively, p<0.002 vs. both ADSC and aSMC).
CONCLUSIONS: Multi-potent ADSCs are readily obtained from subcutaneous fat, grow rapidly, but express low levels of SMC-specific markers at baseline. Switching to a low-serum DM significantly shifts ADSC phenotype toward that of control aSMC but, in contrast to previous reports, no additional shift was seen with growth factors or mechanical stretch. These observations support the potential for SMC-like cells differentiated from ADSC to provide a less invasive source of autologous SMC for clinical application of blood vessel bioengineering. Further research is needed to determine the capacity of these cells to populate vascular scaffolds and the fate of cells after implantation.