Effects of extracellular matrix analogues on primary human fibroblast behavior

2.1. Preparation of gels and sponges

Matrigel^™ (BD, Biosciences Discovery Labware, Bedford, MA) gels were made according to the manufacturer’s protocol. Briefly, the material was thawed overnight on ice at 4°C. Subsequently, the product was kept on ice and handled with cold pipettes. After casting, the material was allowed to gel for 10 min at room temperature in the hood before adding medium.

PuraMatrix^™ (BD, Biosciences Discovery Labware) was prepared according to manufacturer’s instructions. The peptide mix was bath-sonicated for 30 min prior to use. Sterile sucrose solution (10% w/v) was used to dilute the material twofold. The protocol instructions were followed to obtain gels. Briefly, at 5 min after casting, culture medium was gently added on the material to increase the pH of the matrix and promote gelation. The medium was changed three more times during the next 30 min.

PureCol^™ (Inamed, Freemont, CA) was mixed with 10× phosphate-buffered saline (PBS) as recommended by the supplier and the pH of the solution was adjusted to 7.4 with 1 M NaOH. The prepared collagen solution was then filtered through a 0.45 μm syringe driven filter unit prior to casting, to ensure sterility.

Extracel^™ hydrogels (Glycosan BioSystems, Inc., Salt Lake City, UT) were obtained by mixing 1% w/v CMHA-S solution (Glycosil) with 1% w/v Gtn-DTPH solution (Gelin-S) in a 1:1 volume ratio and cross-linking this mixture with 2% w/v PEGDA (mol. wt 3400) in a 4:1 volume ratio. All components were dissolved in DMEM/F12 + 10% newborn calf serum + 2 mM L-glutamine + penicillin/streptomycin (T31 fibroblast growth medium), and the pH of the CMHA-S and Gtn-DTPH solutions were adjusted to 7.5 with 0.1 M NaOH. All solutions were then filtered through a 0.45 μm syringe driven filter unit prior mixing to ensure sterility. The hydrogels were cast and allowed to gel in the hood at room temperature before adding medium.

2.2. Pseudo-3-D cell proliferation assay

T31 human tracheal scar fibroblasts were a generous gift from Dr. S.L. Thibeault, Department of Surgery, University of Wisconsin, Madison, WI). T31 fibroblasts [29,35] are sensitive, non-immortalized cells derived from primary culture and were selected as being representative primary cells. The T31 cells were evaluated under pseudo-3-D plating conditions in 96-well plates. Each material tested (50 μl well⁻¹) was used to coat one row per plate (a total of seven plates were used, one for each day of the assay). The hydrogels were allowed to gel and were seeded with 3.5 × 10⁴ cells ml⁻¹. On the third day of culture, the medium was refreshed. Cell numbers were monitored each day by using the Cell-Titer 96 Aqueous One Solution Cell Proliferation assay (MTS assay) (Promega, Madison, WI). The A₄₉₀ values, which are directly proportional to the number of viable cells, were plotted against the time course of the assay to yield the growth profile of the cells seeded on various sECMs.

2.3. 3-D cell culture

To determine the growth of T31 fibroblasts encapsulated in various gels, cells were entrapped in different materials at a final concentration of 10⁵ cells ml⁻¹ and 100 μl of cell + sECM mix was cast per well of a 24-well plate containing Corning Transwell permeable supports (inserts) with 8.0 μm membrane pore size (Corning Inc., Corning, NY). After gelation, cells were incubated with DMEM/F12 + 10% newborn calf serum + 2 mM L-glutamine + penicillin/streptomycin for 72 h and the MTS assay was used to estimate the number of viable cells in each material.

2.4. Live/dead cell assay

The cytocompatibility of sECMs with T31 fibroblasts was evaluated by analyzing cell viability by a double staining procedure that uses calcein AM and ethidium homodimer-1 (EthD-1) (LIVE/DEAD^® Viability/Cytotoxicity Kit, Invitrogen, Carlsbad, CA). Calcein AM is a nonfluorescent, cell-permeant molecule that is cleaved inside the cell by intracellular esterases to yield its fluorescent counterpart (green fluorescence). EthD-1 is a nucleic acid stain impermeant to viable cells but can diffuse through the membrane of dead cells where it binds to the DNA and gives a red fluorescence. Cells (3 × 10⁵ cells ml⁻¹) were encapsulated in various sECMs and cultured for 6 days (the medium was replaced with fresh one on the third day of culturing). The staining solution used was obtained by diluting a stock solution of calcein AM to a final concentration of 2 μM in a 4 μM EthD-1 solution in 1× PBS, pH 7.4. The inserts with gels were washed three times in 1× PBS, pH 7.4, then incubated for 45 min at room temperature, on a rocker, with the staining solution. After 45 min, the samples were washed again three times in 1× PBS, pH 7.4, then covered with one drop of VectaShield mounting medium for fluorescence (Vector Laboratories, Inc., Burlingame, CA) to prevent photo-bleaching. Cells were analyzed with an Olympus IX70 microscope equipped with Microfire/QCAM CCD (Olympus America Inc., Melville, NY) at ×100 magnification using DAPI and GFP filters.

2.5. Subcutaneous tissue growth in vivo

Twelve 6-week-old female athymic nude mice (Charles River, Wilmington, MA) were anesthetized by intraperitoneal injections of ketamine (80 mg kg⁻¹) and xylazine (10 mg kg⁻¹) according to the protocol approved by the University of Utah Institutional Animal Care and Use Committee (IACUC). The dorsal skin was then sterilized with iodine and alcohol swabs.

T31 fibroblasts were trypsinized and resuspended in 2.5% CMHA-S and 3% Gtn-DTPH in a 1:1 volume ratio. One volume of 4% PEGDA solution was then added to four volumes of cell suspension, and the resulting suspension was mixed gently by vortexing. The final cell concentration was 5×10⁷ cells ml⁻¹. All solutions for the in vivo injections were prepared in serum-free growth medium. When the cell suspension was becoming viscous (~3–5 min at 20°C), 100 μl of gel was injected subcutaneously bilaterally into the back of each mouse by means of an 18-gauge needle for a total of six mice. Another six mice received the bilateral injection of same number of cells in Matrigel^™ according to the manufacturer’s protocol. The concentrations of CMHA-S and Gtn-DTPH were increased compared with the in vitro protocols to support a higher cell seeding density.

Animals were euthanized at 4 weeks post-injection in a CO₂ chamber, and the implants with the surrounding tissues were removed from the mice, fixed in 10% buffered formalin solution (Sigma, Milwaukee, WI)) for 24 h, embedded in paraffin, cut into 5 μm sections, mounted onto slides, and stained with Hematoxylin and Eosin B (H&E; Fisher Scientific, Hanover Park, IL) following standard protocols. Cellular densities in histological slides from both sECM-derived tissues were microscopically counted at identical magnifications.

2.6. H&E staining

Same culturing conditions and experimental duration were used as described under Section 2.4. The hydrogels were removed from the inserts and fixed for 5 min in methanol. Samples were then further processed via H&E staining at the LDS Hospital Electron Microscopy Laboratory (Salt Lake City, UT). The slides were then analyzed with an Olympus IX70 microscope equipped with Microfire/QCAM CCD (Olympus America Inc., Melville NY) at ×400 magnification.

2.7. Statistical analysis

Values, represented as mean ± standard deviation (SD) were compared using Student’s t-test (2-tailed) with p < 0.05 considered statistically significant and p < 0.005 or p < 0.001 considered highly significant.

3.1. Pseudo-3-D proliferation and morphology

First, the proliferation rates and morphology of fibroblasts cultured on thin layers of different biomaterials or tissue culture plate (TCP) were analyzed. We refer to biomaterial-coated culturing conditions as “pseudo-3-D” because the thin layer of biomaterial provides a metabolizable substratum of variable compliance and composition that cells can penetrate and degrade, while the tissue culture plates used for conventional 2-D cell culture present cells with a stiff, non-metabolizable, impenetrable surface.

Two major morphological patterns were observed: stretched, spindle-like shaped fibroblasts and clustered rounded cells. Cells grown on Matrigel^™ adopted a rounded, clustered morphology. In contrast, cells grown on surfaces such as TCP (control), PureCol^™, PuraMatrix^™ or Extracel^™ had a more spindle-shaped appearance (Fig. 1). This experiment was repeated three times with consistent results, showing that the observed differences in cellular morphology were clearly attributable to variations in biomaterial nature and composition.

Consistent with the aforementioned results, cell proliferation rates of fibroblasts cultured on various hydrogels were also different (Fig. 2). All statistical results are reported relative to the TCP control. Matrigel^™ and PureCol^™ sustained a slower cell division rate when compared with cells grown on TCP (control) (p < 0.001). Cells grown on Extracel^™elicited similar proliferation rates with those on TCP, up to day 6. On day 7, cell numbers appeared higher on Extracel^™ than on TCP (p < 0.001), most probably due to 2-D growth surface limitation of the TCP vs. the psedo-3-D surface of Extracel^™. Cells grown on PuraMatrix^™ behaved similarly to the control group grown on TCP (p > 0.05). This was caused primarily by the inconvenient preparation protocol, which causes an incomplete surface coating with PuraMatrix^™; as a result, many cells actually attached to the TCP instead of to the biomaterial.

We did not further characterize T31 fibroblast behavior in pseudo-3-D environments, as the aim of this experimental section was to underscore the importance of transitioning from 2-D to 3-D culture systems for accurate in vivo mimicry. Therefore, we cannot speculate on the different spreading trends on sECMs, since the mechanistic data needed to support any such assumptions were not the object of this study.

3.2. 3-D cell culture

In order to assess cell proliferation rates under 3-D conditions, T31 fibroblasts were encapsulated in hydrogels, as shown in Fig. 3. Three days after seeding, the number of cells in each biomaterial was determined by colorimetric assays. The variance for Matrigel^™ was high due to its rapid gelation at room temperature, with the increasing viscosity of the material causing some pipetting inconsistencies. Therefore, when compared with Matrigel^™, the composition of hydrogels appears to have a minimal, statistically non-significant influence on cellular properties (p > 0.05 relative to Matrigel^™; Fig. 4). A small difference in cellular proliferation rates was observed between PuraMatrix^™ and Extracel^™ (p < 0.05). However, during encapsulation of cells in PuraMatrix^™, problems similar to those in the pseudo-3-D assays were encountered. Cells cultured in PureCol^™ settled to the bottom of the inserts because of the prolonged gelation time of the material (45–60 min at 37°C). Thus, this condition was further excluded from the assay because it failed to mimic a true 3-D growth environment.

3.3. Cytological analysis

The viability and morphology of encapsulated T31 fibroblasts was determined by a double staining procedure that stains live cells green and dead cells red. Three-dimensionally cultured cells were examined 6 days after seeding. This experimental end time-point was arbitrarily chosen based on the assumption that during this time the embedded cells have completely adapted to their new microenvironment. Consistent with the results described above, the composition of biomaterials influenced cell morphology and behavior (Fig. 5). Matrigel^™-encapsulated cells appeared to be spread, with fibroblasts eliciting spindle-shaped morphology (Fig. 5A). As already mentioned, PuraMatrix^™ and PureCol^™ allowed the gravitational settling of cells onto the insert membrane. Cells embedded in both of these biomaterials, grown under pseudo-3-D conditions, had a spindle-shaped morphology (Fig. 5B and C). In Extracel^™, cells were three-dimensionally distributed, but were mostly round-shaped (Fig. 5D). Cell viability in all hydrogels was estimated to be 85–90% based on the double staining microscopic evaluation.

The classical approach of quantifying cell viability in 3-D was not feasible here, since there are multiple vertical and horizontal planes in the 3-D microenvironment that could have led to erroneous interpretations. Thus, Figs. 4 and 5 offer complementary information. Fig. 4 provides a quantitative interpretation of the viability of cells embedded in hydrogels, while Fig. 5 presents additional information on the 3-D embedded cell distribution and morphology.

Next, embedded cells were stained with H&E for a more detailed cytological analysis (Fig. 6). Fibroblasts embedded in Matrigel^™ or Extracel^™ are organized in clusters that correspond to the structures observed by fluorescent staining. PuraMatrix^™ and PureCol^™encapsulated cells appear isolated and concentrated in just a few areas of the hydrogel cross-sections. These observations are consistent with the fact that most of the cells embedded in these materials settled on the bottom of the inserts, resulting in attachment and growth on the insert membrane.

3.4. In vivo subcutaneous injections

Based on our preliminary in vitro evaluations, Matrigel^™ and Extracel^™ were the only two biomaterials suitable for injections into immunocompromised mice. To this end, T31 fibroblasts mixed with either Matrigel^™ or Extracel^™ were injected subcutaneously into nude mice and the quality of the newly formed tissue was assessed. This experiment was modeled on the successful growth of healthy fibrous tissue in vivo from T31 cells cultured on an earlier formulation of the sECM [29]. Both matrices were enclosed in a thin layer of fibrous tissue that delimitated them from the surrounding native tissues. We could thus confirm that the injected fibroblasts were able to survive and proliferate, and that endogenous cell migration did not occur, or occurred to an insignificant extent.

Differences in cell densities between the two matrices are clearly visible in the histological sections (Fig. 7). Fig. 7A shows the histological analysis after the injection of T31 fibroblasts in Matrigel^™ after 4 weeks in vivo. Very little degradation of the ECM equivalent has occurred, and some newly formed immature fibrous tissue is evident but with low cell density. In contrast, 4 weeks after subcutaneous injection of T31 fibroblasts in Extracel^™ (Fig. 7B), the hydrogel has been almost completely degraded and remodeled into newly formed fibrous tissue, with cell density and ECM deposition resembling that of native fibrous tissue.