These authors contributed equally to this work.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

Electrospun nanofibrous structures provide good performance to scaffolds in tissue engineering. We measured the local diffusion coefficients of 3-kDa FITC-dextran in line patterns of electrospun nanofibrous structures fabricated by the direct-write electrospinning (DWES) technique using the fluorescence recovery after photobleaching (FRAP) method. No significant differences were detected between DWES line patterns fabricated with polymer supplied at flow rates of 0.1 and 0.5 mL/h. The oxygen diffusion coefficients of samples were estimated to be ~92%–94% of the oxygen diffusion coefficient in water based on the measured diffusion coefficient of 3-kDa FITC-dextran. We also simulated cell growth and distribution within spatially patterned scaffolds with struts consisting of either oxygen-permeable or non-permeable material. The permeable strut scaffolds exhibited enhanced cell growth. Saturated depths at which cells could grow to confluence were 15% deeper for the permeable strut scaffolds than for the non-permeable strut scaffold.

The goal of tissue engineering is tissue regeneration. Tissue-engineering scaffolds provide the three-dimensional (3D) space necessary for cells to grow into tissues and to then guide their development into specific shapes [

Electrospun nanofibers have high surface-to-volume ratios and a structure similar to that of the extracellular matrices of the human body. Electrospinning techniques use a variety of materials, including natural materials such as collagen and alginate. These characteristics enable good performance in terms of cell adhesion; therefore, a number of recent studies have investigated electrospun scaffolds [

The permeability of struts is another advantage of patterned electrospun scaffolds. The permeability of a “scaffold” is determined by the arrangement and volume fraction of struts, whereas the permeability of the “struts” is determined by the properties of the strut material. A scaffold with permeable struts is expected to improve mass transfer to the inner regions of the scaffold, especially when cells fill the spaces between the struts. Limited mass transfer into the scaffold interior is directly linked to necrosis of internally located cells, a phenomenon observed in most scaffolds developed to date. Necrosis typically occurs when oxygen cannot diffuse further into the interior of a scaffold where cells proliferate [

However, studies of scaffolds have focused mainly on production and fabrication methods, whereas research on diffusion phenomena has received significantly less attention. Therefore, the current study used the fluorescence recovery after photobleaching (FRAP) method to examine microscale diffusion through patterns constructed by electrospinning [

The ladder-patterned nanofibrous structure fabricated using DWES is shown in

_{p}), the scaffold samples were prepared with PCL at flow rates of 0.1 mL/h (_{p} = 0.5 mL/h and 1.007 ± 0.381 μm at _{p} = 0.1 mL/h.

_{p} = 0.1 and _{p} = 0.5 line patterns and 2% calcium alginate hydrogel. Five to seven curves obtained from three to five samples were averaged for each curve. The position of the bleaching spot was changed each time to obtain a single recovery curve. As shown in _{p} = 0.1 mL/h sample and 2% calcium alginate hydrogel were slightly faster than that of the _{p} = 0.5 mL/h sample. However, the differences were not statistically significant (

where _{1/2} is the half-time of recovery and ^{−6} cm^{2}/s) was reasonably consistent with other reports (1.48 × 10^{−6} cm^{2}/s for 4-kDa FITC-dextran [^{−6} cm^{2}/s for 9.4-kDa FITC-dextran [

The oxygen diffusion coefficient was roughly estimated from the diffusion coefficient of 3-kDa FITC-dextran using

where _{0} is the diffusion coefficient of a test molecule in water, _{v}_{s} is the radius of the test molecule, and _{f} is the radius of the fiber. This equation describes the effect of barriers formed by randomly distributed long molecular fibers on the diffusion of a molecule [^{−5} cm^{2}/s [^{−5} cm^{2}/s) is in good agreement with that reported previously (2.54 × 10^{−5} cm^{2}/s at 30 °C [

_{s}/_{w} = 0.93) and non-permeable (_{s}/_{w} = 0) struts. The saturation time, _{s}, was defined as the time required to attain steady state for the _{s}/_{w} = 1 model (~28 days). The cell densities of both permeable and non-permeable cases reached a predetermined maximum value after 0.54 _{s} at the top surface of the scaffold that was directly exposed to fresh media. However, cell density at the bottom surface peaked at 0.13 _{s} and 0.23 _{s} for the non-permeable and permeable strut scaffolds, respectively; moreover, the maximum cell density was much lower than that of the top surface. After peaking, the cell density decreased gradually and reached a steady state. Cell densities were saturated at 98 μm for the permeable strut (_{s}/_{w} = 0.93) scaffold and at 86 μm for the non-permeable strut scaffold. _{s}. The maximum saturated depth was ~99 μm; this value was obtained when the oxygen diffusion coefficient in the strut was identical to that in water, which simulated the situation when the strut region was just filled with water with no obstacles. In the case of the scaffold constructed with a strut diffusion coefficient of 0.75_{w}, the saturation depth was similar to the maximum saturated depth, as shown in

_{s} = 0.93_{w}, a–d) and non-permeable struts (e–h), respectively. In the case of the non-permeable strut scaffold, the cell density distribution was not altered markedly according to the strut arrangement. Additionally, as the volume fraction of the strut (α) increased, the depth to which cells grew to confluence decreased, albeit not to a marked extent.

However, the cell density distribution changed significantly with the strut volume fraction in the case of the permeable strut model. The penetration depth increased as the strut volume fraction increased because the strut region acts as an oxygen-diffusion pass and cell-free region in which oxygen is not consumed.

The maximum thicknesses of the scaffolds in this study differed according to the strut width and height and the pore size. However, our analyses demonstrate that the permeable strut scaffold, such as when fabricated using the DWES method, is advantageous for the transfer of materials and for cell growth. We previously reported several preliminary cell culture tests using patterns or scaffolds fabricated by DWES electrospinning [

Numerous studies of scaffolds have been published; however, few have measured the cell density distribution within scaffolds according to their thickness. Therefore, we have to date been unable to verify our simulation results. Nevertheless, Dunn ^{−6}–4 × 10^{−5} cm^{2}/s. They estimated the critical size of cell-seeded scaffold that can be cultured

PCL (Sigma, St. Louis, MO, USA) was used for electrospinning. PCL (

We used 3-kDa FITC-dextran (Invitrogen, Carlsbad, CA, USA) as a fluorescent probe. The scaffold was soaked in a FITC-dextran solution (100 mg/mL) under a vacuum for ~5 min to allow the solution to penetrate the pattern. The fluorescence signal was stable prior to bleaching. An inverted microscope (Olympus, Tokyo, Japan) with a 20× objective lens was used to form the bleaching spot and capture recovery.

The simulation was performed using the hypothetical 3D scaffold model shown in _{s}. The “cell culture domain”, simulates the space between the struts; cells are assumed to be cultured only in this domain. The height of each strut was set to 50 μm; therefore, the height of each layer of the scaffold was 50 μm. The mesh pattern was created by rotating each layer 90° and stacking layers on top of each other. The total thickness was set to 1000 μm. The diffusion coefficient of the cell culture domain was assumed to be the oxygen diffusion coefficient in typical tissue culture system (_{t}) [_{s} was altered from 0 to 2.68 × 10^{−5} cm^{2}/s; that is, the diffusion coefficient of oxygen in water (_{w}) [_{0}) [

This equation was derived based on oxygen consumption in spaces within which cells were present. The initial and maximum cell densities were set at 2.1 × 10^{11} cell/m^{3}[^{14} cell/m^{3}[

where _{g} is doubling rate, _{g,max} is the maximum cell population doubling rate and _{d} is minimum cell division time [

The effect of the strut volume fraction (α = volume of the strut/total scaffold volume) and arrangement were also investigated. Applied strut arrangements were chosen to be of the “lattice type” (

The diffusion characteristics in a spatially patterned scaffold constructed of a permeable strut were investigated. The diffusion coefficients of fluorescent dyes within the DWES line patterns and alginate hydrogel film were measured using FRAP. We observed no significant differences in diffusion coefficients among the alginate hydrogel and DWES line patterns fabricated with polymer supplied at a flow rate of 0.1 and 0.5 mL/h. The oxygen diffusion coefficient was estimated to be ~92%–94% of the diffusion coefficient of water.

A diffusion simulation was performed in the various scaffold models using the estimated oxygen diffusion coefficient. The saturated depths to which cells could grow to confluence were 86 μm for the non-permeable strut scaffold and 99 μm for the permeable strut (_{s}/_{w} = 1) scaffold when the strut volume fraction was 50%. Therefore, the permeable strut scaffold could be made ~15% thicker.

We found good diffusion within both the electrospun scaffold and the alginate hydrogel. Therefore, the tissue mass regenerated by bioengineering, which to date has been restricted due to mass transfer of adequate gas and nutrients, might be increased by developing our approach further.

Scanning electron microscopy (SEM) micrographs and fiber diameter distribution charts of the nanofibrous structures. (_{p} = 0.1 mL/h (_{p} = 0.5 mL/h (_{p} = 0.1 mL/h (_{p} = 0.5 mL/h (

Results of fluorescence recovery after photobleaching (FRAP) measurements. (

Time based simulation results for the quantity of cells in the various scaffold models. Two points (bottom and top) for electrospun line patterns (_{s}/_{w} = 0.93) and for the non-diffusion strut model (_{s}/_{w} = 0).

Saturated depth according to strut diffusion coefficient.

Cell density distribution at _{s}. (_{s} = 0.93 _{w}) strut with a lattice-type arrangement. α = 70% (_{s} = 0.93 _{w}) strut with a staggered type arrangement. α = 50%; (

Schematic diagram of the direct-write electrospinning (DWES) apparatus.

Fluorescence recovery after photobleaching (FRAP) system for diffusion coefficient measurements. (

Schematic representation of the simulation domain. (

Parameter values and definitions.

Parameter | Definition | Value | Ref. |
---|---|---|---|

_{max} |
Maximum cellular oxygen consumption rate | 3.3 × 10^{−16} mol/cell/s |
[ |

_{m} |
Half-maximum rate oxygen concentration | 3.79 × 10^{−3} mol/m^{3} |
[ |

Saturation constant in Monod kinetics | 3 nmol/mL | [ | |

_{0} |
Maximum dissolved oxygen concentration | 0.1 mol/m^{3} |
[ |

_{t} |
Oxygen diffusion coefficient in typical tissue culture sysyem | 2.0 × 10^{−5} cm^{2}/s |
[ |

_{s} |
Oxygen diffusion coefficient in scaffold strut domain | 0–2.68 × 10^{−5} cm^{2}/s |
- |

_{d} |
minimum cell division time | 36.5 h | [ |

This research was supported by the Pioneer Research Center and Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (NRF-2010-0004324, NRF-2012-0009666, 2012R1A1A1015738), and the Human Resources Development Program (20114010100070) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry and Energy, Republic of Korea.

The authors declare no conflict of interest.