br The differential scanning calorimetry
The differential scanning calorimetry (DSC) thermograms of the PLLA electrospun fibers (A-PLLA and R-PLLA) exhibit the existence of the free amorphous region as revealed by the appearance of the recrystallization peak in the temperature range of 70e90 C. In
Fig. 1. Morphological comparison of MDA-MB-231 ABT888 cultured on different substrates under normoxic and hypoxic conditions for 7 days. Immunofluorescence of breast cancer cells was imaged with Hoechst 33342 (blue), phalloidin (green), and E-cadherin antibody (red). Scale bar: 100 mm. Arrows indicate the aligned fiber direction of the substrate. PLLA, poly(L-lactic acid); PCL, poly(ε-caprolactone).
where Ef is the elastic modulus of the electrospun fiber substrate measured by the stressestrain curve. The aligned fibers have significantly stronger tensile proper-ties in comparison to random electrospun fibers. The random orientation of the fibers improves the ultimate strain at break in each mat created by electrospinning. The fiber orientation during
contrast, for PCL fibers, such extra heat absorbed by the recrystal-lization does not show in the profile (data not shown). The characteristics for these fiber substrates derived from FE-SEM micrographs, tensile test, and DSC measurement are summa-rized in Table S2 (Supplementary data). The average fiber diameter of the R-PCL fiber mat exhibits larger value in comparison to that of A-PCL. From stressestrain curve of the tensile test (1.2. Tensile test: Supplementary data), the elastic modulus, fracture stress, and ul-timate strain were assessed. The relative density (rf/rb) corresponds to the density of the electrospun fiber substrate divided by the bulk density. The expected elastic modulus of the bulk substrate (Eb) is related by the following relationship as expected from the defor-mation model proposed by Gibson and Ashby .
the collecting process in an aligned fiber might accelerate the crystallization in the molecular level in an electrospun PLLA fiber.
The porosity values calculated by the buoyancy method (1.2. Tensile test: Supplementary data) are 0.748, 0.712, 0.565, and 0.649 for A-PLLA, R-PLLA, A-PCL, and R-PCL, respectively. The porosity is promoted between PLLA substrates by the aligned process of electrospinning. But the porosity does not exactly follow the same trend in PCL substrates.
3.2. Effect of fiber topography and stiffness on cellular morphology
Fig. 1 shows the cellular morphology of MDA-MB-231, and Fig. 2 shows the cellular morphology of MCF-7 breast cancer cells cultured on different substrates under both normoxic and hypoxic
Ep ¼ Ef (1) conditions at day 7. For comparison, the cellular morphology at day
3 is shown in Fig. S4 (Supplementary data) (MDA-MB-231) and Fig. S5 (Supplementary data) (MCF-7). The corresponding morphological parameters are summarized in Fig. 3 and Fig. 4. For MCF-7 cells incubated on all substrates, CDH1 (red fluorescence) was detectable and localized at the intercellular boundaries (bor-ders), indicating the multicellular aggregates (colonization) of the cells (Fig. 2 and Fig. S5). On the other hand, this type of aggregation was not observed for MDA-MB-231 cells (Fig. 1 and Fig. S4), but the
Fig. 2. Morphological comparison of MCF-7 cells cultured on different substrates under normoxic and hypoxic conditions for 7 days. Immunofluorescence of breast cancer cells was imaged with Hoechst 33342 (blue), phalloidin (green), and E-cadherin antibody (red). Scale bar: 100 mm. Arrows indicate the aligned fiber direction of the substrate. PLLA, poly(L-lactic acid); PCL, poly(ε-caprolactone).
Fig. 3. Quantification of the cellular morphologies of cytoplasm roundness ((a) and (a’)), cytoplasm elongation factor ((b) and (b’)), nuclear elongation factor ((c) and (c’)), and AN/AC ratio ((d) and (d’)) as boxplots for MDA-MB-231 cells cultured under normoxic and hypoxic conditions at day 7. PLLA, poly(L-lactic acid); PCL, poly(ε-caprolactone).
cells were elongated and arranged along the fiber direction when they were incubated on the aligned fibers substrates.
For MCF-7 cells, the aligned fiber substrates (A-PLLA) produce less elongation and alignment of the cells along the fiber orienta-tion directions. This is because MDA-MB-231 cells exhibit a more mesenchymal nature with weaker cellecell adhesion than MCF-7 cells [32e34]. It is well demonstrated that both cell alignment
and migration with a guide are introduced by line topographies (such as fibers or microgrooves) [21,35e37].
In particular, for MCF-7 cells cultured on A-PCL, the effect of alignment of the substrate on the cellular morphology was barely observed (Figs. 2 and 4(b), (c), (b’), and (c’)). This phenomenon was similar to the results reported by Saha et al. . The cells could spread and adhere to the stiff substrate rather than the soft sub-strate [15,18,38]. Thus, the combination of both surface topogra-phies and stiffness of the substrate should be used to evaluate the effects on the cellular morphology.