Differentiation of Adipose-Derived Stem Cells into Vascular Smooth Muscle Cells for Tissue Engineering Applications

Abstract

Synthetic grafts have been developed for vascular bypass surgery, however, the risks of thrombosis and neointimal hyperplasia still limit their use. Tissue engineering with the use of adipose-derived stem cells (ASCs) has shown promise in addressing these limitations. Here we further characterized and optimized the ASC differentiation into smooth muscle cells (VSMCs) induced by TGF-β and BMP-4. TGF-β and BMP-4 induced a time-dependent expression of SMC markers in ASC. Shortening the differentiation period from 7 to 4 days did not impair the functional property of contraction in these cells. Stability of the process was demonstrated by switching cells to regular growth media for up to 14 days. The role of IGFBP7, a downstream effector of TGF-β, was also examined. Finally, topographic and surface patterning of a substrate is recognized as a powerful tool for regulating cell differentiation. Here we provide evidence that a non-woven PET structure does not affect the differentiation of ASC. Taken together, our results indicate that VSMCs differentiated from ASCs are a suitable candidate to populate a PET-based vascular scaffolds. By employing an autologous source of cells we provide a novel alternative to address major issues that reduces long-term patency of currently vascular grafts.

Keywords: adipose stem cells; contraction; differentiation; tissue engineering; vascular smooth muscle.

Figure 1

TGF-β and BMP-4 induces differentiation of human ASCs into SMC-like phenotype. (A) Top are representative immunoblots for αSMA, caldesmon and SM22α expression. Corresponding line graph demonstrate the time-course (0 to 7 days) effect of TGF-β and BMP-4 on αSMA, caldesmon and SM22α expression. * p < 0.05 vs. αSMA expression at day 0, ** p < 0.05 vs. SM22α expression at day 0, # p < 0.05 vs. caldesmon expression at day 0. (B) Fluorescence microscopy was also used to detect αSMA, caldesmon and SM22α expression in hASCs at day 0, 4 and 7 of TGF-β and BMP-4 stimulation. Scale bar 100 µm. (C,D) Contractile activity of non-differentiated hASCs (Day 0) and hASCs differentiated for 7 days with TGF-β and BMP-4 embedded into collagen gel matrices. The collagen gel matrices were cultured in the serum-free medium for up to 6 h and gel contraction was photographed at the indicated time points using a digital camera. Corresponding line graph demonstrating the time-course (0 to 6 h) effect of 60 mM of KCl on the area of the gel lattices relative to its original size. * p < 0.05 vs. non-differentiated hASCs at 6 h, ** p < 0.05 vs. non-differentiated hASCs at 6 h. All results are mean ± SEM of 5 experiments.

Figure 2

TGF-β and BMP-4 have a synergistic effect on human ASC differentiation. hASCs were stimulated either with TGF-β, BMP-4 or in combination (DM) for up to 7 days and αSMA (A), caldesmon (B) and SM22α (C) expression were evaluated by Western blot and fluorescence microscopy (D). Scale bar 100 µm. Top are representative immunoblots for αSMA, caldesmon and SM22α expression. Corresponding line graph demonstrating the time-course (0 to 7 days) effect of TGF-β, BMP-4 or DM on αSMA, caldesmon and SM22α expression. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. day 0 of stimulation with DM; ** p < 0.05 vs. day 0 of stimulation with TGF-β; # p < 0.05 vs. day 0 of stimulation with BMP-4.

Figure 3

IGFBP7 expression is increased in human ASCs going through differentiation with TGF-β and BMP-4. (A) Top is representative immunoblot for IGFBP7 expression. Corresponding line graph demonstrating the time-course (0 to 7 days) effect of TGF-β and BMP-4 on IGFBP7 expression. TGF-β and BMP-4 synergistically induce the expression of IGFBP7 on hASC. hASCs were stimulated with TGF-β, BMP-4 or combination of TGF-β and BMP-4 (DM) for up to 7 days and IGFBP7 expression was evaluated by Western blot. Top are representative immunoblots for IGFBP7 expression. Corresponding bar graph demonstrating the time-course (0 to 7 days) effect of TGF-β, BMP-4 or DM on IGFBP7 expression. (B) Fluorescence microscopy was also used to detect TGF-β, BMP-4 or DM-induced IGFBP7 expression on day 0, 4 and 7. Scale bar 100µm. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. day 0 of stimulation with DM; ** p < 0.05 vs. day 0 of stimulation with TGF-β; # p < 0.05 vs. day 0 of stimulation with BMP-4.

Figure 4

Combination of IGFBP7 and BMP-4 but not IGFBP7 alone induced expression of molecular markers of SMC. Human ASCs were stimulated with differentiation media containing TGF-β and BMP-4 (DM), IGFBP7 or combination of IGFBP7 and BMP-4 for up to 7 days. Top are representative immunoblots for αSMA (A), caldesmon (B) and SM22α (C) expression. Corresponding line graph demonstrating the time-course (0 to 7 days) effect of DM, IGFBP7 or IGFBP7 and BMP-4 on αSMA, caldesmon and SM22α expression. (D) Western blot analysis was further confirmed by fluorescence microscopy to detect αSMA, caldesmon and SM22α expression at day 0, 4 and 7 following stimulation with DM, IGFBP7 or IGFBP7 and BMP-4. Scale bar 100 µm. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. day 0 of stimulation with DM; ** p < 0.05 vs. day 0 of stimulation with or IGFBP7 and BMP-4.

Figure 5

TGF-β- and BMP-4-induced differentiation of human ASCs is stable for up to 21 days. (A) Top are representative immunoblots for αSMA, caldesmon and SM22α expression. Corresponding bar graph demonstrating the stability of differentiation when cells were differentiated for 7 days and then either maintained in differentiation media containing TGF-β and BMP-4 or transferred to DMEM supplemented with FBS for indicated times. (B) Fluorescence microscopy was also used to detect αSMA in hASsC differentiated for 7 days and either kept in DM or switched to DMEM supplemented with FBS for indicated times. Scale bar 100 µm. (C) Contractile activity of non-differentiated hASCs and hASCs differentiated for 7 days and then either maintained in differentiation media containing TGF-β and BMP-4 or transferred to DMEM supplemented with FBS for indicated times. The collagen gel matrices were cultured in the serum-free medium for up to 6 h and gel contraction was photographed at the indicated time points using a digital camera. Corresponding line graph demonstrating the time-course (0 to 6 h) effect of 60 mM of KCl on the area of the gel lattices relative to its original size. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. non-differentiated ASC.

Figure 6

Human ASCs can be differentiated into SMC after 4 days of treatment with TGF-β and BMP-4. (A) Rather than 7 days, cells were differentiated for 4 days and then transferred to DMEM containing FBS for indicated times. αSMA and caldesmon expression was detected by fluorescence microscopy. Scale bar 100 µm. (B) Contractile activity of non-differentiated hASCs and hASCs differentiated for 4 days and then transferred to DMEM supplemented with FBS for indicated times. The collagen gel matrices were cultured in the serum-free medium for up to 6 h and gel contraction was photographed at the indicated time points using a digital camera. Corresponding line graph demonstrating the time-course (0 to 6 h) effect of 60 mM of KCl on the area of the gel lattices relative to its original size. Bar graphs indicate the area of the gel lattices relative to its original size after 6 h. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. non-differentiated ASC.

Figure 7

PET structures do not affect the differentiation of human ASCs into SMC mediated by TGF-β and BMP-4. Long-term stability of differentiation was evaluated on hASCs grown in PET surface. Cells were differentiated for 7 days and then either maintained in differentiation media containing TGF-β and BMP-4 or transferred to DMEM supplemented with FBS for indicated times. (A) Top are representative immunoblots for αSMA, SM22α and caldesmon expression. Corresponding bar graph demonstrate protein expression of αSMA and caldesmon evaluated by Western blot. (B) Fluorescence microscopy was also used to detect αSMA, SM22α and caldesmon expression in ASCs grown in PET surface at indicated time points. Scale bar 100 µm. Results are mean ± SEM of 5 experiments. * p < 0.05 vs. non-differentiated ASC.