Fibrous hydrogel scaffolds with cells embedded in the fibers as a potential tissue scaffold for skin repair
- 812 Downloads
A novel approach was undertaken to create a potential skin wound dressing. L929 fibroblast cells and alginate solution were simultaneously dispensed into a calcium chloride solution using a three-dimensional plotting system to manufacture a fibrous alginate scaffold with interconnected pores. These cells were then embedded in the alginate hydrogel fibers of the scaffold. A conventional scaffold with cells directly seeded on the fiber surface was used as a control. The encapsulated fibroblasts made using the co-dispensing method distributed homogeneously within the scaffold and showed the delayed formation of large cell aggregates compared to the control. The cells embedded in the hydrogel fibers also deposited more type I collagen in the extracellular matrix and expressed higher levels of fgf11 and fn1 than the control, indicating increased cellular proliferation and attachment. The results indicate that the novel co-dispensing alginate scaffold may promote skin regeneration better than the conventional directly-seeded scaffold.
The mild gelling conditions of alginate solutions allow cells to be encapsulated inside alginate hydrogel and remain viable. These cells have been shown to proliferate, differentiate, and express their cellular products [1, 2, 3, 4, 5, 6]. In their study, Takei et al.  enclosed bovine carotid artery vascular endothelial cells in alginate hydrogel fibers and used them as templates for endothelialized tubes. By embedding cells in certain materials, the growth rate can be manipulated by changing the concentration of the polymer and initial cell-seeding density .
Most studies used cells encapsulated in alginate gel beads whose sizes ranged from hundreds of micrometers [1, 8] to several millimeters [3, 5] and were intended for cell therapy use. Abbah et al. , encapsulated mouse adipose tissue stromal cells in alginate hydrogel beads (1.5 % cross-linked in 200 mM calcium solution). Those cells were cultivated in osteogenic medium for 21 days and showed higher viability, alkaline phosphatase activity, osteocalcin expression, and calcium secretion compared to cells cultured in a monolayer. Several studies have also found that cell morphology, the proliferation rate, and metabolic activity are related to the concentrations of alginate and calcium crosslinker [2, 3, 5]. Lee et al.  demonstrated that differentiation was dominant in osteoprogenitor cells cultured in rigid low-density beads with a small mesh size while proliferation was dominant in compliant low-density beads with a larger mesh size. Hunt et al.  reported that fibroblast cells encapsulated in alginate gel beads accelerated the degradation process of the gel, possibly because the calcium ions remaining within the alginates were used by the cells and thus not available for cross-linking with the alginate.
In addition to being used in cell therapy, alginate hydrogel is a popular biomaterial used in making scaffolds for tissue repair because of its excellent biocompatibility [9, 10, 11, 12]. To the authors’ best knowledge, in these research reports, cells were always seeded/grown on the surface of the alginate hydrogel within the pores of the tissue scaffolds. The characteristics of cells embedded inside the alginate hydrogel of a porous scaffold in terms of cell morphology, proliferation, and cellular expression have yet to be investigated. We tested how cells proliferate and express their cellular activities when grown inside the alginate hydrogel fibers of a scaffold constructed using the three-dimensional (3D) plotting method. Alginate scaffolds with cells seeded on the fiber surface were used as controls. The morphological changes of the cells and their distribution within the scaffold over time were investigated. After skin trauma, fibroblasts gather at the injury site and quickly proliferate and secrete extracellular matrix (ECM), mostly type I collagen and fibronectin (fn1), to close the injury. To assay these cellular activities during skin wound repair, cell proliferation, viability, and type I collagen deposition in the extracellular matrix were quantified, as well as the expression of genes regulating cell growth and attachment to the ECM, namely fibroblast growth factor (fgf11) and fn1.
2 Materials and methods
2.1 Alginate scaffold manufacturing by 3D plotting
2.2 Cell seeding
Fibroblasts (L929, ATCC, VA, USA) were cultured in DMEM supplemented with 10 % fetal bovine serum, 100 U/mL of penicillin, and 100 μg/mL of streptomycin (all culture medium supplements were obtained from Invitrogen, CA, USA). The cells were kept in a CO2 incubator (SCA-165DS, ASTEC, Fukuoka, Japan) at 37 °C with 95 % humidity and 5 % CO2.
Co-dispensing In total, 5 mL of cell suspension in DMEM containing a total of 2.5 × 106 cells was mixed with 15 mL of 4 % w/v alginate/DMEM solution so that the final alginate concentration was 3 % w/v and the final cell concentration was 1.25 × 104/ml. Each scaffold was composed of 200 μl of the above mixture and therefore contained approximately 25,000 cells embedded in its fibers.
Conventional direct seeding Alginate scaffolds were made using cell-free 3 % w/v alginate/DMEM solution as described in “Alginate scaffold manufacturing by 3D plotting”. The scaffolds were then rinsed in DMEM several times to remove CaCl2 and then placed in the wells of a 24-well tissue culture plate. Excess DMEM was removed from each well, and 20 μl of cell suspension (containing 25,000 cells) was placed on top of each scaffold. The scaffolds were placed in a shaker incubator for 1 h at 95 rpm to ensure homogeneous cell seeding on the scaffold.
All scaffolds were placed in a CO2 incubator for up to two weeks, and cell culture medium was changed every other day. The cells were then tested at 1, 4, 7, and 14 days after cell seeding.
2.3 Sample morphology
2.3.1 Light microscopy and H&E staining
After the scaffolds were constructed, random samples were selected, and their individual fibers were inspected using a light microscope (CK X41, Olympus, Shinjuku, Japan) to ensure that there was no excessive bubble entrapment in the fibers and that the cells were homogeneously distributed inside the fibers.
After the cells were cultured for 1, 4, 7, and 14 days, the samples were rinsed with phosphate-buffered saline (PBS, pH = 7.2) and cross-linked in 2.5 % v/v glutaraldehyde/PBS solution for 30 min. The samples were then rinsed with PBS and dehydrated in ethanol solutions ranging from low to high concentration. The samples were kept in 99.5 % (v/v) ethanol before testing. The fiber layers were carefully separated from each other before being placed in Mayer’s hematoxylin solution (Sigma, MO, USA) for 5 min. The fibers were rinsed with DI water for 5 min and differentiated in acid ethanol (125 μL HCl + 50 mL 70 % (v/v) ethanol) for 30 s before being placed in eosin Y solution (0.1 in 90 % ethanol) for 5 min. The samples were rinsed with DI water before being observed under a light microscope.
2.3.2 Scanning electron microscope (SEM)
As described in “Light microscopy and H&E staining”, the samples were cross-linked and dehydrated with ethanol. Hexamethyldisilazane (Sigma) was used to remove the residual ethanol from the samples before they were air dried. To make cross-sections, dry samples were placed in liquid nitrogen for 1 min and carefully cut open using a scalpel. The specimens were sputter coated with gold–palladium (E1010, Ion Sputter, Hitachi, Tokyo, Japan) before being examined using scanning electron microscopy (SEM, S-3000H, Hitachi).
2.4 Cell viability and three-dimensional distribution
Cell viability and distribution within the scaffold were qualitatively assessed using the Live/Dead® viability stain (Molecular Probe, OR, USA) according to the manufacturer’s directions. Briefly, the samples were washed with PBS and incubated with 200 μL of LIVE/DEAD® reagent (2 mM calcein AM + 4 mM ethidium homodimer-1) for 30 min at 37 °C. The cells were then observed using a laser confocal microscope (TCS-SP5-AOB5, Leica, Solms, Germany). Viable cells appeared green while dead cells appeared red. The 3D scanning function of the microscope was used, and images taken at different depths were captured and overlapped to reconstruct the three-dimensional distribution of cells in the sample.
2.5 Cell proliferation
Each sample (n = 6) was incubated in 1 mL of 100 mM EDTA for 30 min to dissolve the hydrogel scaffold and release the cells. The cell suspension was centrifuged at 3,000 rpm for 10 min, and the supernatant was discarded. A total of 2 mL of 1 % papain solution (0.1 w/v % papain in 0.1 M Na2HPO4 + 5 mM Na2EDTA, pH = 6) was added to the cells and kept at 55 °C for 1 h. The DNA content in the digest was then measured using a commercial kit (Quant-iT™ PicoGreen, Invitrogen, CA, USA). A luminescence spectrometer (LS 55, PerkinElmer Instruments, MA, USA) was used to determine the fluorescent intensity (Ex/Em = 480/530 nm) of the cell digest. The intensity was then converted to total DNA concentration (ng/mL) using a standard curve.
2.6 Type I collagen production in the ECM
The amount of type I collagen secreted in the ECM was measured using a Sircol™ Collagen Assay (Biocolor, UK) following the manufacturer’s instructions. Prior to the test, each sample (n = 6) was incubated in 1 mL of 100 mM EDTA for 30 min, and 2 mL of pepsin solution (0.1 mg/mL 0.5 M acetic acid) (Sigma, MO) was added to each sample which were kept at 4 °C overnight. The acidic extract was then moved to a microcentrifuge tube, and 100 μL of neutralizing reagent and 200 μL of isolation reagent were added. The mixture was incubated overnight at 4 °C, and the tube was centrifuged at 12,000 rpm for 10 min. The supernatant was discarded, and 1 mL of Sircol Dye was added to the precipitate. The contents were gently mixed for 30 min, and the tube was then centrifuged at 12,000 rpm for 10 min. The supernatant was removed, and the pellet was rinsed with 750 μl of acid-salt wash. The tube was again centrifuged at 12,000 rpm for 10 min, and all liquid was removed. Alkali reagent (250 μl) was added to the tube, which was then vortexed for 5 min before the absorbance of the liquid was read at 555 nm.
2.7 Real-time PCR
Total RNA was isolated using the RNeasy kit (Qiagen, CA, USA) following the manufacturer’s instructions. The isolated total RNA (1 μg) was reverse transcribed into cDNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA).
Specific primers for PCR amplification listed with expected fragments size
PCR product length (bps)
Real-time PCR was performed using a real-time PCR system (StepOnePlus, Applied Biosystems) and a commercially available kit (Power SYBR® Green PCR Master Mix, Applied Biosystems). The reaction volume was 20 μl. The reaction mixture contained 10 μl of mix reagent, 1 μl of forward primer (5 μM), 1.0 μl of reverse primer (5 μM), 2 μl of cDNA template, and 6 μl of DEPC water. After an initial cycle at 95 °C for 10 min to activate the polymerase, each of the remaining 38 cycles consisted of 15 s at 95 °C, 30 s at 45 °C for GAPDH, 50 °C for Col 1a1, 48 °C for fn1, and 47 °C for fgf11, and 60 s at 60 °C. A cycle threshold value (Ct) was obtained for each sample (n = 6). The 2−ΔΔCt method was used to calculate the relative expression of each target gene.
2.8 Statistical analysis
All data are presented as the averages ± standard deviation. Statistical analysis was performed using a one-way ANOVA with Tukey’s test to compare the differences in means between co-dispensing and direct seeding groups (SigmaStat 3.5 for Windows, Systat Software, Chicago, IL, USA). A difference of P < 0.05 between co-dispensing and direct seeding cells and fibroblast expression over time was considered to be statistically significant.
3.1 Scaffold and cell morphology
3.1.1 Light microscopy (H&E staining)
Directly-seeded cells were more densely populated in localized areas and therefore began to form aggregates as early as day 4 (arrow, Fig. 2f). Additionally, on day 4, the processes of directly-seeded cells could be observed (circle, Fig. 2f) while co-dispensed cells still appeared circular. Directly-seeded fibroblasts began to form larger aggregates on day 7 (arrow, Fig. 2h) while co-dispensed cells proliferated and remained homogeneously distributed within the fibers. By day 14, the directly seeded cells attached to the alginate surface displayed the typical fibroblastic spindle shape (arrow, Fig. 2j) with the majority of cells in clusters. Co-dispensed cells within the fibers formed elongated aggregates on day 14 (Fig. 2i). We believe that the cells had propagated in stretched air bubbles or fractures within the gel. Additionally, as the gel degraded over time in the culture, more space became available, enabling the growth of aggregates within the gel . Similar cell proliferation patterns have been observed in alginate gel discs and beads before. Fibroblasts encapsulated in alginate discs and beads were initially found to distribute homogeneously within the hydrogel matrix and began to form clusters after two weeks in culture as the gel matrix degraded [2, 20].
The cell-electrospinning technique has also been shown to be an effective method of cell therapy, building complex living structures, and fabricating various tissue scaffolds [21, 22, 23]. A coaxial needle arrangement with a cell suspension flew through the inner needle and a poly(dimethylsiloxane) medium with high viscosity and low electrical conductivity flew through the outer needle was first used to fabricate microfibers containing living cells [22, 23]. Cells were cultured post-electrospinning and shown to be viable. These encapsulated living cells from the electrospinning process were released into culture in a matter of days after the encapsulating polymer dissolved. This left little time for cells to lay down ECM and form a biological structure. Jayasinghe  fabricated living vessel architectures by cross electrospinning cell-containing fibers on to a rotating mandrel. It resulted in an uniform distribution of cells throughout the fabricated structure with segregated cell types at different layers in the tubular construct. Live cell electrospinning, however, was found to cause high cell death rates (up to 60 %) and required the careful selection of cell-friendly polymers suitable for electrospinning . Hu et al.  described another method of fiber spinning for transporting live cells in the fibers. Cell-seeded solids and hollow hydrogel fibers were produced by enzymatically cross-linking gelatin-hydroxyphenylpropionic acid in solutions flowing within a capillary tube. The majority of the embedded cells from this mild fiber formation process were viable.
3.1.2 SEM (top and cross-section views)
The patterns of cell proliferation and aggregate formation observed in the SEM results were similar to those observed in the H/E test. Other studies have also reported similar cell proliferation patterns. Pokrywczynska et al.  mixed murine 3T3 fibroblasts at different cell densities (2.0−6.0 × 106 cells/ml) with 1 % alginate and cross-linked the mixture using CaCl2 to form disc-shaped scaffolds. They showed that cell spheroids formed within the discs after two weeks in culture. Cells encapsulated in alginate beads form aggregates after approximately two weeks of culture, and higher concentrations of alginate delay the time of aggregate formation [2, 5].
Rapidly proliferating cells have been shown to form spheroids more quickly than those that proliferate more slowly in vitro . Because mammalian cells do not have receptors for and do not readily adhere to unmodified alginate , they tend to form aggregates during rapid proliferation . During tissue repair, large aggregate formation is not generally desirable because necrotic regions form within the spheroids due to insufficient nutrient and oxygen perfusion to their centers.
3.2 Cell viability and spatial distribution of cells
This cell encapsulation study was performed on a much larger scale than previous studies which focused on encapsulating cells in beads. With a 3D plotting system, cells can be homogeneously distributed in scaffolds with 100 % pore connectivity, allowing efficient nutrient/metabolic waste transportation in vivo. With the conventional direct-seeding method, cells distributed on one side of the fibers in a localized area, and large cell aggregates appeared as early as day 7. These aggregates were large enough to block the scaffold pores and thus the transportation of nutrients/metabolic waste. For in vivo applications, the existence of pores is also important for angiogenesis which usually begins four days after skin injury . It has been reported that pores of larger size and higher interconnectivity and porosity support rapid and extensive angiogenesis . Expedited angiogenesis in porous scaffolds is known to reduce hypoxia and improve the survival of transplanted cells . The delayed aggregate formation observed in the co-dispensing method is favorable for angiogenesis and prevents early cell necrosis within the aggregate before vessel formation.
3.3 DNA quantification (proliferation)
From the perspective of cell–cell interaction, cadherins in the cell membrane can form complexes with growth-factor receptors and modulate cell activation and/or stability at the cell membrane. When cells reach confluence, growth factor receptors may preferentially associate with cadherins and inhibit proliferation (contact inhibition) [32, 33, 34]. The directly-seeded cells were more concentrated on the surface of the scaffold near where they were initially seeded. They entered the stage of confluence/contact inhibition earlier than the co-dispensed cells, which were scattered within the fibers. Therefore, on day 4 and day 7, the growth of directly-seeded cells was slightly lower than that of co-dispensed cells (P < 0.05). The crowded directly-seeded cells began to grow on top of one another and formed aggregates. The cells were able to form large aggregates within the pores without experiencing the same mechanical restriction as the aggregates did in the co-dispensed scaffolds. Additionally, the co-dispensed cells began to experience the effect of contact inhibition and the higher mechanical constraints of the hydrogel  by day 14. Therefore, the amount of co-dispensed cells was almost 25 % less than directly-seeded cells.
3.4 Type I collagen expression in the ECM
Vascular endothelial growth factor (VEGF) mediates proliferation , and proliferating cells produce ECM, predominantly type I collagen. In sparse endothelial cells, vascular endothelial (VE) cadherin is diffuse on the cell membrane. In confluent cells, VE-cadherin is clustered on the membrane at junctions, and VEGF induces survival (via apoptosis prevention) instead of cell proliferation. We hypothesized that only cells on the periphery of the aggregates were actively dividing and producing type I collagen. Those within the aggregate experienced contact inhibition and switched the cellular mechanism from collagen production to apoptosis prevention . Therefore, on day 14, the majority of directly-seeded cells were metabolically inactive due to contact inhibition, and collagen was produced only by the cells on the periphery of the aggregates.
Another important factor contributing to the lower ECM collagen content in directly seeded scaffolds was that the collagen produced could not accumulate on the alginate surface because negatively charged procollagen molecules cannot be retained by the anionic alginate gel . Additionally, the enzyme BMP-1, which is associated with collagen fibril production, is also negatively charged at physiological pH . In contrast, in co-dispensed models, collagen fibrils can be produced and trapped inside the small spaces between cell clusters and the surrounding gel. Smith et al.  postulated that cellular calcium requirements cause localized degradation of the alginate around cell clusters and that within this space, it is likely that a small amount of collagen begins to assemble as procollagen, and BMP-1 are secreted by the cells.
3.5 Real-time PCR of fgf1 and fn1
The expression of fn1 was higher in the co-dispensed group than in the directly-seeded group from day 4 to day 14 (P < 0.05) (Fig. 7b). In both groups, fn1 expression increased from day 1 to day 4 (P < 0.05) and declined from day 4 to day 14 (P < 0.05). Fibroblasts secrete fn1 to aid in adhesion to their substrate . The majority of the cells in the co-dispensed group were surrounded by gel on the initial days of culture (Fig. 2c, e, g) and on later days, were surrounded by the type I collagen they accumulated in the pericellular spaces . It is logical to assume that compared to the directly-seeded cells, higher fn1 expression was stimulated in co-dispensed cells because of their more extensive contact with the surrounding substrate. Additionally, fibroblast growth factor 2 (fgf2) has been reported to upregulate the expression of fn1 , and it is possible that the higher degree of fgf11 expression in co-dispensed cells may have stimulated higher fn1 expression. The directly-seeded cells were unable to accumulate significant amounts of ECM collagen on the alginate surface, and the cells within the aggregates were in survival mode and were not expected to express fn1.
An alginate scaffold with fibroblasts encapsulated in its hydrogel fibers was successfully constructed, and its performance as a potential skin repair scaffold was tested. Gel-encapsulated fibroblasts distributed more evenly in the scaffold and showed delayed formation of large cell aggregates compared to cells that were directly seeded onto the scaffold. Delayed aggregate formation suggests a higher degree of angiogenesis within the scaffold, resulting in more efficient nutrient/waste exchange for the cells in later stages of wound healing. The encapsulated cells in the scaffold also deposited more ECM type I collagen and expressed higher levels of fgf11 and fn1 than the directly-seeded cells which indicates higher cellular activities of proliferation and attachment. These results indicate that alginate dressings with cells encapsulated in the fibers could better promote skin regeneration.
In a previous study performed in this laboratory, this same type of scaffold with dimensions of 6 cm × 6 cm × 0.2 cm was constructed without cells . It was easily handled without breaking. It is reasonable to believe that the cell-embedded scaffold made in this study could be manufactured in sizes practical for clinical use.
This study was funded by the National Science Council in Taiwan (NSC 100-2221-E-027-026) and the National Taipei University of Technology and Taipei Medical University Joint Research Program (NTUT-TMU-101-08).
- 2.Bohari SP, Hukins DW, Grover LM. Effect of calcium alginate concentration on viability and proliferation of encapsulated fibroblasts. Biomed Mater Eng. 2011;21(3):159–70.Google Scholar
- 6.Smith AM, Hunt NC, Shelton RM, Birdi G, Grover LM. Alginate hydrogel has a negative impact on in vitro collagen 1 deposition by fibroblasts. Biomacromolecules. 2012;13(12):4032–8.Google Scholar
- 12.Khanna O, Larson JC, Moya ML, Opara EC, Brey EM. Generation of alginate microspheres for biomedical applications. J Vis Exp. 2012;66:e3388.Google Scholar
- 29.Nguyen DT, Orgill DP, Murphy GF Chapter 4: The pathophysiologic basis for wound healing and cutaneous regeneration. biomaterials for treating skin loss. Cambridge/Boca Raton: Woodhead Publishing (UK/Europe) & CRC Press (US); 2009 pp. 25–57.Google Scholar
- 31.Singh S, Wu BM. Dunn JC. J Tissue Eng Regen Med: Enhancing angiogenesis alleviates hypoxia and improves engraftment of enteric cells in polycaprolactone scaffolds; 2012.Google Scholar