A scaffold with a bio-mimetically designed micro/nano-fibrous structure using decellularized extracellular matrix

Abstract

Decellularized extracellular matrix (d-ECM)-based scaffolds have been extensively applied in various tissue regeneration applications because they regulate various cell functions and effectively guide new tissue formation. However, fabrication methods using d-ECM have been limited by its low processability. In this study, a new fibrous scaffold consisting of poly(ε-caprolactone) (PCL) and d-ECM was fabricated using an electrohydrodynamic jet process to obtain a pore-controlled multi-layered structure. In the scaffold, the d-ECM was used as a supplementary bioactive component to induce highly active cell responses. The suggested PCL/ECM fibrous structure showed significantly higher tensility (tensile modulus: 2-fold) than a pure PCL fibrous structure with a similar pore structure. The in vitro cellular responses of the fibrous structure were increased using human fibroblasts, and the ECM-based scaffold showed significantly higher cell-seeding efficiency (1.8-fold) and metabolic activities (1.5-fold at seven days) than pure PCL with a similar pore size and porosity. These results suggest that the d-ECM-based scaffold is promising as a biomedical substrate to effectively regenerate tissues and that this fabrication method will be very useful for designing biomimetic biomedical scaffolds.

---

Introduction

Numerous attempts have been made to obtain biomedical scaffolds that are highly effective at regenerating various tissues by using decellularized natural extracellular matrix (d-ECM) components.^1–5 The natural ECM, which consists of a complex micro-fibril physical structure and various proteins, has several functions; it provides physical support to cells, supplies various bioactive cues that modulate cell proliferation and differentiation, and produces a flexible physical environment enabling vascularization and new tissue formation by tissue dynamic processes.⁶ For these reasons, the best scaffolds for tissue regeneration may involve the native ECM, but exact mimics of the highly complex structure and functions of the ECM can be extremely difficult to reproduce. Therefore, various scaffolds that simply mimic the functions and structure of the natural ECM by using natural and synthetic biopolymers are a realistic alternative.

Three-dimensional (3D) scaffolds mimicking the native ECM using biomaterials have been fabricated in line with the following basic concepts: they should have a porous structure, allowing the transport of nutrients, metabolic waste, and growth factors; appropriate mechanical properties; controllable biodegradability; and micro/nanostructured surface topography, among others.^6–8 In addition, it has been widely reported that controlling the physical structure, including microspores and porosity, of scaffolds and biochemical compounds to stimulate cells within these scaffolds can regenerate various tissues.⁹

Natural biomaterials, such as protein-based biomaterials (collagen, elastin, fibrinogen, actin, gelatin, etc.) and polysaccharide-based biomaterials (alginate, dextran, cellulose, chitin/chitosan, glucose, etc.)¹⁰ usually provide excellent biocompatibility to enhance cell attachment and proliferation. However, these biomaterials have several shortcomings, such as limited mechanical stability; significantly low processability to obtain realistic 3D complex shapes, such as that of a human organ, and to control micro/nano pore structures; and high immunogenicity, particularly for xenograft proteins. In contrast, synthetic biomaterials such as poly(glycolic acid), poly(lactic acid), poly(lactide-co-glycolide), poly(ε-caprolactone) (PCL), and polyurethane exhibit controllable mechanical properties, good processability, and relatively low immunogenicity, but their low biocompatibility for the attachment and growth of cells has been a major issue to overcome. To overcome the shortcomings of both natural and synthetic biomaterials, various composite and blending systems using appropriate combinational processing techniques,^11–13 including conventional leaching methods, solid-freeform fabrication, electrospinning, and a simple coating/dipping method, have been applied to achieve tailor-made biomimetic scaffolds for the regeneration of particular tissues.

The newly developed electrohydrodynamic jet (EHDJ) process in a wet state has been recently used as an efficient processing technique to obtain biomimetic nonwoven micro-fibrous multi-layered structures, which consist of macroscale bundles of micro/nanofibers and macro pores, both over 100 μm.^14,15 The EHDJ process used ethanol solution (EtOH) as a target medium to generate threads consisted of microsized fibers. The mechanism for the EHDJ process was explained in previous work.¹⁶ Concisely, the fibrous bundles consisting of the micro/nanofibers in the pure ethanol were obtained due to the enforced breakup of the electrically charged single jet in the ethanol. This phenomenon was occurred due to rapid interchanging from the high surface tension of the solvent of the PCL solution with the relatively low surface tension of the ethanol in the bath.¹⁶ The fabricated mesh-like structure consists of multi-layered macro-cylindrical struts assembled with entangled micro/nanofibers. The controllable macro pore size and high porosity of the fabricated structures allow the efficient transport of nutrients and metabolic waste between the cell-seeded mesh structure and the local environment, and also provide a realistic 3D mesh structure consisting of micro/nanofibers having high surface-to-volume ratios for the effective delivery of biochemical cues to cultured cells. This unique structure can overcome the submicron pore size of the conventional 2D electrospun mat, which can constrain the rapid growth and migration of cultured cells in the thickness direction of the mat, by accommodating a layer-by-layer arrangement of macroscale struts.

Here, we suggest a new biomedical scaffold fabricated using the EHDJ process that combines a synthetic material (PCL) with two different concentrations of d-ECM to overcome the following limitations: the low ability to process/shape d-ECM and the low biocompatibility of PCL. To avoid structural deficiencies of electrospun fibrous scaffolds, we used the EHDJ process to control the pore structure (pore size and porosity) in the multi-layered fibrous scaffold. The scaffolds were analyzed not only in terms of their surface morphology and physical properties—including mechanical properties, fibrous morphology, and dynamic water uptake ability—but also their biological activities—such as cell-seeding efficiency and metabolic activities—using human dermal fibroblasts (HDFs) and compared with a pure PCL fibrous scaffold with a similar pore structure as a control fabricated using the same process.

Experimental

Materials

Poly(ε-caprolactone) (PCL) [density: 1.135 g cm⁻³; molecular weight: 90 000 g mol⁻¹; melting point: 60 °C] was purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). To fabricate nanofibers, 10 weight fractions of PCL in a 20 : 80 solvent mixture of methylene chloride (surface tension: 28.1 mN m⁻¹) and dimethylformamide (surface tension: 37.1 mN m⁻¹) (Junsei Chemical Co., Tokyo, Japan) were used with a 16-G electrospinning nozzle and a 20 mL glass syringe. In the EHDJ process, 99% ethanol (EtOH; surface tension: 22.1 mN m⁻¹; Duksan, South Korea) was used in the target bath. Adult HDFs from adult human foreskin (MCTT Co., South Korea) were used in this work.

Decellularization of the cellular ECM on an electrospun fibrous mat and freeze-milling

For cell culture, we used electrospun PCL fibers that were fabricated for 10 min (Fig. 1a). To decellularize the cell-cultured fibrous mat, we followed the protocol of Chen et al.¹⁷ After culturing cells on the fibrous mat for 4 days and 4 weeks, the cell-cultured fibrous structures were washed with phosphate-buffered saline and MilliQ water. Freeze–thaw cycling with NH₄OH aqueous solution was used for the decellularization process, and the cultured samples were frozen at −80 °C in water for 3 h, thawed at room temperature, and then washed with purified water.¹⁷ The freeze–thaw cycle was repeated 6 times, and the specimens were immersed in an aqueous solution of ammonia (25 mM) for 20 min on a slowly moving four-way shaker. Finally, to remove the ammonia, the specimens were washed 6 times. The decellularized PCL/ECM was freeze-dried for 24 h, and the PCL/d-ECM was freeze-milled into a fine powder for use in the EHDJ process (Fig. 1b).

Fig. 1 (a) Schematic of electrospinning and cell seeding of human dermal fibroblasts on the micro/nanofibers. (b) Freeze-milling process for fabricating PCL/d-ECM powder. (c) Three PCL-based solutions of 10 wt% PCL and PE solutions, consisting of 10 wt% PCL and two different weight fractions of d-ECM powders. (d) Fabrication of 3D structure by EHDJ process with the PCL solutions containing d-ECM powder.

The DNA content in the cultured cells on the fibrous mat was determined fluorometrically using PicoGreen assay kit (Life Technologies). Cells were lysed using TE solution (10 mM Tris-HCL, 1 mM EDTA, pH 7.5) at 37 °C for 2 h. Samples of 100 μL were placed in triplicate in a 96-well plate and mixed with 100 μL of Quant-iT PicoGreen reagent. The plate was incubated at room temperature for 10 min in the dark and then read on fluorescent micro-plate reader (Synergy HT; Bio-Tek Instruments, USA) using standard fluorescein wavelengths (excitation/emission wave length: 480 nm/520 nm). Five samples were tested during the incubation period, and each test was performed in triplicate.

Preparation of PCL/ECM ink

Three kinds of solutions for fabricating scaffolds were prepared in this study and were referred as PCL, PE-1, and PE-2, respectively. 20 g of 10 weight fraction of pure PCL solution was prepared by melting 2 g of PCL in 18 g of MC (80): DMF (20) solvent mixture. 1 g of PCL/d-ECM powder and 1 g of PCL were mixed in the solvent solution, instead of 2 g of pure PCL, to make 20 g of PE-1. PE-2 solution was prepared by melting 2 g of PCL/d-ECM powder in 18 g of the solvent (Fig. 1c).

ECM components (glycosaminoglycan (GAG) and collagen) quantification

The GAG content of the specimens was determined using the Blyscan Sulfated Glycosaminoglycan Assay Kit (Biocolor Ltd., Newtownabbey, Republic of Ireland). Samples were prepared by digesting each construct using papain extraction reagent (0.1 mg mL⁻¹ papain) and heated at 65 °C for 3 h. After centrifugation at 10 000 × g for 10 min, the supernatant was collected and assayed in accordance with the manufacturer's protocol.

The collagen contents of samples were determined by colorimetric analysis applying a hydroxyproline assay kit (Sigma-Aldrich Inc., St. Louis, MO, USA), following the protocol from the manufacturer. 10 mg of the samples were homogenized in 100 μL of 3-distilled water and hydrolyzed in 12 M HCl at 120 °C for 3 hours in capped vials. 100 μL of supernatant was then mixed with 200 μL of assay reagent, and incubated at 60 °C overnight, followed by transfer of 100 μL of the sample reaction mixture into a 96-well plate for absorbance readings at 560 nm. The collagen contents of samples were then computed based on an assumption of a collagen-to-hydroxyproline ratio of 10 : 1 (w/w).

Fabrication of a fibrous structure using EHDJ

Several electric fields and a fixed flow rate (1.5 mL h⁻¹) were used in the EHDJ process. The electric field system was connected to a multi-axis robot moving system (DTR3-2210 T-SG; DASA Robot, South Korea). The speed of the nozzle connected to the robot was set to 5 mm s⁻¹. In this process, the flow rates of the pure PCL and PCL/d-ECM solutions were controlled using a syringe pump (KDS 230; KD Scientific, Holliston, MA, USA). Electric field strength was set using a power supply (SHV300RD-50K; Convertech, Seoul, South Korea), and the EHDJ processing temperature and humidity were fixed at 28 °C and 35 ± 4%, respectively. In addition, the height of EtOH in the target bath was fixed at 4 mm. More details of the effects of the EHDJ processing parameters on the formation of fibrous bundles have been described in our previous papers.^14,15 After direct printing of the fibrous bundles in the target medium, they were washed with pure water and freeze-dried (SFDSM06; Samwon, Busan, South Korea) at −76 °C for 2 days (Fig. 1d).

Sample characterization

Fibronectin from cells cultured on the fibrous structures was fixed in paraformaldehyde (3.8%) for 30 min and permeabilized with 0.05% Triton X-100 for 15 min at room temperature. The specimens were blocked in PBS containing 0.2% bovine serum albumin (BSA) for 1 h at 25 °C. The primary antibody (fibronectin, ab6328; Abcam) was diluted 1 : 200 and incubated with the sample for 1 h at 25 °C. An Alexa Fluor® 488-conjugated donkey anti-mouse antibody, diluted 1 : 400, was used as the secondary antibody.

Scanning electron microscopy (SEM; SNE-3000M; SEC, Inc., South Korea) was used to observe the fibrous morphology and pore size. Sample preparation and measurements were performed in accordance with the manufacturer's instructions. The porosity (%) of the fibrous structures was obtained using the equation [1 − M/(ρV)] × 100, where M represents the mass of the fibrous scaffold, ρ is the density of the PCL, and V is the volume of the fibrous structure, which was assumed to be rectangular. The geometry of the scaffold was measured using a digital caliper micrometer. In this work, we assumed that the density of the PCL/d-ECM mixture was the same as that of pure PCL because the weight fraction of the embedded d-ECM was very small relative to the weight fraction of PCL.

The uptake of water by the fibrous scaffolds was determined by weighing the scaffolds before and after soaking in distilled water for 12 h. The increase in water absorption (%) was calculated as (W_{12 h} − W₀)/W₀ × 100, where W_{12 h} is the weight of the scaffold after 12 h and W₀ is the weight at time zero.

To obtain tensile stress–strain curves, the samples were cut into small strips (8 × 20 mm²). Uniaxial tests were carried out using a tensile machine (Top-tech 2000; Chemilab, South Korea). The stress–strain curves for the fibrous mats were recorded at a stretching speed of 0.5 mm s⁻¹. All values are expressed as the mean ± standard deviation (SD; n = 5).

In vitro cell culture and analysis

The fibrous structures (PCL, PE-1, and PE-2) (8 × 10 mm) were sterilized with 70% EtOH and ultraviolet (UV) light, and then placed in culture medium overnight. The HDFs were cultured in fibroblast growth medium (FGM) containing 2% fetal bovine serum (FBS), 0.1% insulin, recombinant human fibroblast growth factor-B (rhFGF-B), and 0.1% GA-1000 antibiotic solution (Lonza, USA). Cells from passage 1 were used for subsequent experiments. Adherent HDFs were rinsed thoroughly with Dulbecco's phosphate-buffered saline (DPBS; WelGENE, South Korea) and detached using 0.05% trypsin–EDTA (Gibco, USA) prior to seeding onto the mats. Cells were seeded onto the specimens at a density of 1 × 10⁵ per sample and incubated at 36 °C in an atmosphere of 5% CO₂. The medium was changed every 3 days.

To determine the cell-seeding efficiency, we used the measuring protocol of Sobral et al.¹⁸ Briefly, the seeded cells were left in the scaffolds for 12 h to provide them with sufficient time to attach. After 12 h, the scaffolds were removed and the cells remaining in the wells were counted. The efficiency of each scaffold was calculated by taking into account the initial number of cells seeded and the residual number of cells in the respective wells after 12 hours. Five specimens of each scaffold were used. The seeding efficiency (%) was calculated as (cells added to scaffold − cells in wells)/(cells added to scaffold) × 100.¹⁸

The proliferation of viable cells was determined using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), a yellow tetrazole (Cell Proliferation Kit I; Boehringer Mannheim, Germany). In this assay, mitochondrial dehydrogenases of viable cells cleave the yellow MTT substrate to produce purple formazan crystals. Cells in the scaffolds were incubated with MTT (0.5 mg mL⁻¹) for 4 h at 37 °C, and the absorbance at 570 nm was measured using a microplate reader (EL800; Bio-Tek Instruments, USA). Five samples were tested during the incubation period, and each test was performed in triplicate.

After culturing the fibroblasts in fibrous scaffolds for 1 and 7 days, they were exposed to 0.15 mM calcein AM for 45 min in an incubator. The stained scaffolds were then analyzed with a confocal microscope (LSM 800; ZEISS, Germany). To measure cell density, the number of green spots (live cells) was counted using ImageJ software (NIH, Bethesda, MD, USA).

After 18 days of cell culture, the pure PCL, PE-1, and PE-2 scaffolds were subjected to diamidino-2-phenylindole (DAPI) staining to visualize cell nuclei in the scaffold. Phalloidin (Invitrogen) was used to visualize the actin cytoskeletons of proliferated cells in the scaffolds.

Statistical analyses

All experiments were performed with at least three replicates for each sample. Data are expressed as the mean ± SD. Statistical comparisons were performed using student's t-test. In all analyses, p values < 0.05 were considered to indicate statistical significance. NS indicates non-significant.

Results and discussion

Decellularization of HDFs on the PCL nanofibrous mat

Fig. 1a–e shows schematics of the decellularization process to attain a natural ECM using HDFs and the EHDJ process to fabricate multi-layered PCL/d-ECM fibrous bundles. Firstly, the electrospun PCL fibrous mat was prepared to culture the fibroblasts (1 × 10⁵ mL⁻¹) (Fig. 1a). Cells cultured on PCL for 4 days and 4 weeks were decellularized by freeze-drying (Fig. 2b). Next, the PCL fibrous mat and ECM were ground with a freezer-mill and the powder was remixed with PCL solution (final weight fraction of PCL solution, 10 wt%) for use in the EHDJ process (Fig. 1c). To quantify the d-ECM in the mixture, we used GAG as a quantitative standard. Finally, the three solutions (pure PCL and two PCL/d-ECM mixtures having different d-ECM weight fractions) were directly plotted over the EHDJ process (Fig. 1d).

Fig. 2 (a) Fluorescence [diamidino-2-phenylindole (DAPI)/fibronectin/F-actin] images of cells cultured for 4 days and 4 weeks. (b) Fluorescence DAPI staining of nuclei before and after decellularization for 4 day and 4 week culture. (c) Quantification of DNA content before and after decellularization. (d) Optical and scanning electron microscopy (SEM) images of the poly(ε-caprolactone)/decellularized ECM (PCL/d-ECM) powders, which were freeze-dried/freeze-milled.

The fluorescence images presents the amount of cells and ECM with the DAPI/phalloidin staining and the stained fibronectin, respectively, in Fig. 2a. The cells cultured for 4 days did not have enough ECM for the use as the printing ink, while the fibrous mats with cell culture for 4 weeks were almost completely covered by cells and ECM. Therefore, the decellularization process was performed for the PCL fiber mats with the sufficient cell culture for 4 weeks.

As the result of the decellularization, the DNA content remained in the mat was examined and also the fluorescence images in Fig. 2b show the stained nuclei of before and after decellularization for 4 day and 4 week culture, respectively. The DNA content of both 4 day and 4 week cultured samples decreased distinctly as in Fig. 2c after the decellularization process. Especially, the PCL fiber mat cultured for 4 weeks contained 834.7 ± 18.5 ng of DNA per sample before the decellularization procedure, while the DNA content has become below 1 ng (0.7 ± 0.1 ng) after the procedure. The results showed that the decellularization was performed successfully. The decellularized fibrous mats cultured for 4 weeks were then ground into fine powder form (Fig. 2d) via freeze-milling process.

Analysis of ECM components in PCL/d-ECM ink

The prepared powder was mixed with PCL solution as explained in the Experimental section (Fig. 1c), and the amount of ECM components such as GAG and collagen has been measured for the three kinds of ink, which were referred to PCL, PE-1, and PE-2, respectively. No GAG was contained in the pure PCL solution, while 0.11 and 0.22 mg of GAG were in PE-1 and PE-2, respectively, as in Table 1. In addition, collagen contents showed a similar tendency as the d-ECM contents in the PE solutions increased (0, 17 and 37 μg in PCL, PE-1, and PE-2, respectively). The increments of ECM contents in the PCL solutions have led to the electrical conductivity and the surface tension changes, of which the values are presented in Table 1. As a result, the changed conductivity and surface tension of the PCL/d-ECM solutions would be main factors of determination for an appropriate EHDJ condition in the process.

Table 1 ECM components (glycosaminoglycan (GAG) and collagen), electrical conductivity, and surface tension of the pure PCL, PE-1, and PE-2 solutions

	PCL	PE-1	PE-2
Total GAG protein (mg) in 20 g of 10 wt% PCL solution	0	0.11	0.22
Total collagen (μg) in 20 g of 10 wt% PCL solution	0	17	37
Electrical conductivity (μs cm^−1)	1.7	22.7	35.0
Surface tension (N cm^−1)	1.9	2.3	2.6

---

EHDJ process for attaining multi-layered PCL/d-ECM fibrous structures

It has been reported that an ideal 3D scaffold can be a multi-layered structure consisting of microsized struts, which can be constituted of micro/nanofibers, to provide high cellular responses.¹⁹ To date, the ideal scaffold can't be fabricated using various conventional methods. However, the EHDJ process has provided a significant development to realize the 3D multi-layered fibrous scaffold with controlled pore structure.¹⁴ Stable initial jet motion is generally required for a successful EHDJ process. For this reason, we measured the initial jet motion of the three solutions at various electric voltages (4–10 kV) and two different distances (20 and 30 mm) between the nozzle tip and the EtOH solution surface. The EHDJ process is shown schematically in Fig. 1d; an electric field was applied and EtOH solution was used as the target medium. In this process, a charged single jet can be directly immersed in the EtOH and printed by a two-axis moving system (Fig. 1d).

Optical images of the initial jets from the nozzle for the different solutions (PCL, PE-1, and PE-2), the applied electric fields under a fixed flow rate (0.15 mL h⁻¹) of the solution, and the distance (20 mm) between the nozzle and the EtOH surface were shown in Fig. S1(a and b).† As shown in these images, with an increase in the applied voltage, the jet behavior changed from an undeveloped state and a stable state to an unstable state of the initial jet. In addition, for the PE-1 and PE-2 solutions, the stable region of the initial jet was wider than that of the pure PCL due to higher electrical conductivity. However, upon increasing the distance from 20 to 30 mm, PE-1 and PE-2 solutions generated substantial whipping compared with the pure PCL (Fig. S1b†). Through simple processing tests, we were able to select an appropriate processing regime, with an electric voltage of 7 kV and a distance of 20 mm, which can stably fabricate microfibrous bundles during the EHDJ process. Fig. 3a shows single printed fibrous bundles, which were freeze-dried after printing on the EtOH bath, of pure PCL, PE-1, and PE-2. As shown in the SEM images, micro/nano-sized fibrous bundles were well fabricated for all solutions. Fig. 3b illustrates the results of the fibronectin staining in the pure PCL, PE-1 and PE-2 scaffolds. As shown in the SEM images, micro/nano-sized fibrous bundles were well fabricated for all solutions. We stained the fibronectin which is one of the most important components of the ECM,²⁰ in order to show the presence of the d-ECM in the scaffolds. The findings of simple fibronectin staining suggested that the d-ECM was well embedded and distributed in the fabricated fibrous structure (PE-1 and PE-2), and the staining was much larger in the PE-2 scaffold which indicates greater ECM component. As expected, no staining was observed in the pure PCL. However, the non-uniform staining of the fibronectin was observed which was caused by the inhomogeneous distribution of the d-ECM powder in the PCL/d-ECM solution.

Fig. 3 (a) SEM images of single struts of PCL, PE-1, and PE-2 printed in the target bath under the processing conditions (electric field strength: 7 kV; distance: 20 mm; flow rate: 1.5 mL h⁻¹). (b) Fluorescence images of fibronectin (green) showing increasing distribution as the content of PCL/d-ECM powder increases.

Fig. 4a–c shows optical and SEM images of the multi-layered fibrous PCL and PE scaffolds fabricated using the EHDJ process. In the SEM images, all fibrous scaffolds were well composed of the micro-fibrous bundles, and the average pore size and porosity were about 695 ± 52 μm and 88.5%, respectively (Fig. 4d and e). In the fibrous bundles, the average diameters of the constituent fibers were 4.1 ± 1.6 μm for pure PCL, 2.9 ± 1.4 μm for PE-1, and 2.6 ± 1.5 μm for PE-2 (Fig. 4f). These results confirmed that the fabricated scaffolds have a similar pore size and porosity, which were needed in order to compare the in vitro cell tests for the fabricated scaffolds.

Fig. 4 Optical and scanning electron microscopy images of the fabricated mesh structure, which can be used in a scaffold using (a) pure poly(ε-caprolactone) (PCL), (b) PE-1, and (c) PE-2 solutions under the processing conditions (electric field strength: 7 kV; distance: 20 mm; flow rate: 1.5 mL h⁻¹). (d) Pore size, (e) porosity, and (f) fiber diameter of the fibrous bundles for the PCL, PE-1, and PE-2 scaffolds. ‘NS’ means the non-significance.

Tensile properties

The mechanical properties of scaffolds are among the most important parameters for evaluating them because they provide not only physical support, but also promote cellular activities for attachment and proliferation.²¹ According to Discher et al., appropriate mechanical strength of biomedical substrates is essential for cell adhesion and survival after seeding cells.²² Fig. 5a and b shows the tensile stress–strain curves attained using the tensile tester for pure PCL, PE-1, and PE-2. The elastic moduli of the tensile stress–strain curves were obtained using their initial slopes. The pure PCL, PE-1, and PE-2 fibrous structures had Young's moduli of 2.1 ± 0.2, 3.2 ± 0.2, and 4.4 ± 0.4 MPa, respectively. The tensile modulus of the PE-2 scaffold was significantly higher than those of the pure PCL and PE-1 because the mixed d-ECM components (fibril collagen, elastin, GAG, etc.) secreted by cells can mechanically reinforce the micro/nanofibers, eventually resulting in much higher tensility.

Fig. 5 (a) Tensile stress–strain curves and (b) Young's modulus and maximum strength of PCL, PE-1, and PE-2 scaffolds.

Dynamic wetting and water uptake of the fibrous structures

To be implanted, biomedical scaffolds must be able to take up water because this property influences their capacity to both absorb body fluid and to perform a variety of cellular responses (cell attachment and homogeneous proliferation). Fig. 6a–c shows the dynamic water wetting, contact angle, and water uptake ability, respectively. As shown in the optical images of a red droplet mixed with water and rhodamine, the pure PCL fibrous structure provided significantly slower water wetting due to the hydrophobic nature of the fibrous PCL, while the PE-2 structure showed significantly faster wetting. To relatively and quantitatively compare the water wetting abilities, we measured the contact angle of the droplet on the surface, and the results indicated that the PE-2 fibrous structure showed significantly higher spreading of water due to the relatively larger amount of embedded hydrophilic d-ECM components. In addition, the PE-2 scaffold had a markedly greater ability to take up water than the pure PCL and PE-1 fibrous structures, the explanation for which is similar to that for the difference in water wetting.

Fig. 6 (a) Dynamic wetting of a droplet (water and rhodamine: red) for poly(ε-caprolactone) (PCL), PE-1, and PE-2 scaffolds at 6 min. (b) Contact angle of the droplet. (c) Water absorption of PCL, PE-1, and PE-2 scaffolds. Asterisks indicate significant differences.

In vitro fibroblast activities on the PCL/d-ECM fibrous structure

To observe the practicability of the d-ECM-assisted biomedical scaffold, various cellular responses on the fibrous structures were evaluated in vitro with HDFs, which were originally used to obtain the d-ECM. The fibrous structures of pure PCL, PE-1, and PE-2 scaffolds were similar in terms of pore size (the distance between fibrous bundles) and porosity.

Fig. 7a shows the cell-seeding efficiency for the pure PCL, PE-1, and PE-2 scaffolds. The efficiency of initial cell seeding is generally one of the most important factors for scaffold-based regenerative medicine because it is directly related to the loss of invaluable cells. To calculate the cell seeding efficiency of the scaffolds, the remained cells in the culturing well were counted after the cell-seeded scaffolds were removed and the percentage of the remained cells to the initially seeded cells was calculated. As shown by the results, the seeding efficiency levels of the pure PCL, PE-1, and PE-2 were 25.9 ± 1.6%, 41.6 ± 2.5%, and 43.5 ± 2.1%, respectively. This difference could be due not only to the previously measured wetting ability and water uptake capability of the PE scaffolds, but also biological proteins (fibronectin and laminin, etc.), that is, cell surface receptors that can facilitate the attachment of cells to the ECM, although the fibrous physical structures had similar pore sizes and porosities.

Fig. 7 (a) Cell-seeding efficiency indicating a distinct difference between PCL scaffold and PE-1/2 scaffolds. (b) 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay at 1, 3, and 7 days for the PCL, PE-1, and PE-2 scaffolds. (c) Fluorescence images of live cells (green) at 1 and 7 days and (d) live cell density (cell number per mm²). (e) Nucleus (blue)/F-actin (red) images for PCL, PE-1, and PE-2 scaffolds after cell culture for 18 days. Asterisks indicate significant differences and ‘NS’ means the non-significance.

Fig. 7b indicates the cell proliferation of viable cells determined by the MTT assay for the pure PCL and PE scaffolds. As shown in the results, the optical density (OD) of the PE scaffolds was significantly higher than that of pure PCL for all culture periods (1, 3, and 7 days). In addition, the rate of proliferation, which was calculated using a simple linear regression of OD vs. the culture period, was a slope = 0.039 and R² = 94.3% for pure PCL, and a slope = 0.054 and R² = 99.8% for PE-2. Fig. 7c shows live-cell fluorescence images at 1 and 7 days for the pure PCL and PE scaffolds. The number of live cells cultured in PE scaffolds was much greater than for pure PCL (Fig. 7d). These results indicate that the d-ECM-based scaffolds are associated with significantly higher metabolic activities than the pure PCL scaffold.

Immunofluorescence images of fibrous pure PCL and PE scaffolds after cell culture for 18 days are also shown in Fig. 7e. In these images, nuclei (blue) and F-actin (red) are shown on the fibrous bundles. They were distributed more densely on the PE scaffolds than on the pure PCL, demonstrating that the HDFs proliferated more efficiently, particularly in terms of cytoskeletal activity, in the PE scaffolds than in the pure PCL. Based on these results, the PE scaffolds can provide a much better microcellular environment that promotes cellular activities, such as cell adhesion and growth, than the pure PCL fibrous scaffold.

Conclusion

This paper presents a new method to obtain an ECM-based fibrous scaffold consisting of a synthetic polymer and d-ECM. The multi-layered PCL/d-ECM fibrous scaffold was fabricated using the EHDJ process. To clarify the cellular activities of embedded ECM components, two different amounts of d-ECM were mixed in the PCL solution. The ECM-based scaffolds were analyzed with regard to various physical and cellular responses using human fibroblasts. The tensile properties of the scaffold were significantly enhanced compared with those of the pure PCL scaffold due to the reinforcing components of the ECM (i.e., fibril collagen, elastin, and GAG). The in vitro cellular responses of the PCL/d-ECM scaffolds indicated that the embedded ECM in the scaffolds had prominent effects on the fibroblasts in terms of cell-seeding efficiency and metabolic activities. However, from this study, we cannot suggest an optimal range regarding the amount of ECM, but we can show the promise of ECM-based fibrous scaffolds as a potential biomaterial for various tissue regenerative applications and also a fabrication method for obtaining such scaffolds.

Acknowledgements

This study was supported by a grant from the National Research Foundation of Korea grant funded by the Ministry of Education, Science, and Technology (MEST) (Grant no. NRF-2015R1A2A1A15055305).

References

1. T. W. Gilbert, T. L. Sellaro and S. F. Badylak, Biomaterials, 2006, 27, 3675–3683.
2. S. F. Badylak, D. O. Freytes and T. W. Gilbert, Acta Biomater., 2009, 5, 1–13.
3. H. C. Ott, T. S. Matthiesen, S. K. Goh, L. D. Black, S. M. Kren, T. I. Netoff and D. A. Taylor, Nat. Med., 2008, 14, 213–221.
4. J. J. Song and H. C. Ott, Trends Mol. Med., 2011, 17, 424–432.
5. Q. Xing, K. Yates, M. Tahtinen, E. Shearier, Z. C. Qian and F. Zhao, Tissue Eng., Part C, 2015, 21, 77–87.
6. B. P. Chan and K. W. Leong, Eur. Spine J., 2008, 17, S467–S479.
7. D. W. Hutmacher, Biomaterials, 2000, 21, 2529–2543.
8. H. Jeon and G. Kim, J. Mater. Chem. B, 2014, 2, 171–180.
9. S. J. Hollister, Nat. Mater., 2005, 4, 518–524.
10. B. D. Ratner, A. S. Hoffman, F. J. Schoen and J. E. Lemons, Biomaterials science: an introduction to materials in medicine, Academic press, 2004.
11. Y. Zhu, C. Gao, X. Liu and J. Shen, Biomacromolecules, 2002, 3, 1312–1319.
12. B. Duan, X. Y. Yuan, Y. Zhu, Y. Y. Zhang, X. L. Li, Y. Zhang and K. D. Yao, Eur. Polym. J., 2006, 42, 2013–2022.
13. L. Ghasemi-Mobarakeh, M. P. Prabhakaran, M. Morshed, M. H. Nasr-Esfahani and S. Ramakrishna, Biomaterials, 2008, 29, 4532–4539.
14. M. S. Kim and G. Kim, Langmuir, 2014, 30, 8551–8557.
15. M. Kim and G. H. Kim, Chem. Eng. J., 2015, 279, 317–326.
16. G. H. Yang, M. Kim and G. Kim, J. Colloid Interface Sci., 2015, 450, 159–167.
17. H. X. Lu, T. Hoshiba, N. Kawazoe and G. P. Chen, Biomaterials, 2011, 32, 2489–2499.
18. J. M. Sobral, S. G. Caridade, R. A. Sousa, J. F. Mano and R. L. Reis, Acta Biomater., 2011, 7, 1009–1018.
19. G. Kumar, C. K. Tison, K. Chatterjee, P. S. Pine, J. H. McDaniel, M. L. Salit, M. F. Young and C. G. Simon, Biomaterials, 2011, 32, 9188–9196.
20. Q. Xing, K. Yates, M. Tahtinen, E. Shearier, Z. Qian and F. Zhao, Tissue Eng., Part C, 2015, 21, 77–87.
21. M. G. Haugh, C. M. Murphy, R. C. McKiernan, C. Altenbuchner and F. J. O'Brien, Tissue Eng., Part A, 2011, 17, 1201–1208.
22. D. E. Discher, P. Janmey and Y. L. Wang, Science, 2005, 310, 1139–1143.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27845g

‡ These authors contributed equally to this study.

---