In Vitro and Ex vivo Analysis of Hyaluronan Supplementation of Integra ® Dermal Template on Human Dermal Fibroblasts and Keratinocytes

Preparation and seeding of Integra^® scaffolds

Lyophilized sodium hyaluronate (NovaMatrix, Norway) (MW 1,400,000-1,800,000 g mol⁻¹) was dissolved in DMEM (Dulbecco's modified Eagle medium) (PAA, Germany), supplemented with 10% (v/v) fetal bovine serum (FBS), 2 mmol/mL⁻¹ L-glutamine, 100 U/mL⁻¹ penicillin/streptomycin for fibroblast culture and ex vivo culture, or in EpiLife^® media (Invitrogen, UK) for keratinocyte culture, at 0.15 mg/mL⁻¹, 1 mg/mL⁻¹, 1.5 mg/mL⁻¹ and 2 mg/mL⁻¹. Integra^® scaffolds (Integra LifeSciences, USA) were washed with PBS and cut into 7 mm disks for in vitro assays and 4 mm disks for ex vivo assays using a biopsy punch (Steifel, USA). The HA-supplemented media was added to Integra^® to a final volume of 200 μl for 16 hours prior to performing assays (See Supplementary Fig. S1, available online as supplementary material at www.jab-fm.com). Unsupplemented Integra^® (UI) was soaked in complete DMEM or EpiLife^®.

Primary human dermal fibroblasts (PHDFs) were obtained from the skin of the lower abdomen of three Caucasian males and two female patients undergoing elective abdominoplasty surgery with full ethical approval [North West Research Ethics Committee (Ref11/NW/068)] according to previously established protocols (27) (See supplementary Methods available online at www.jab-fm.com). Human adult epidermal keratinocytes (HEKa) cells were obtained from Invitrogen, UK).

Generation of ex vivo artificial cutaneous wounds

Full thickness lower abdominal skin, obtained as above (Ref11/NW/068), was trimmed of excess adipose tissue and 8 mm punches taken. In the center of these cylindrical skin biopsies, a 4 mm full thickness artificial wound was created by a further punch biopsy. Models were placed into 24-well plate inserts (Corning, USA) and prepared Integra^® inserted into the wound area. Complete DMEM was added so that the epidermis of the ex vivo tissue was air exposed (Supplementary Fig. S2).

In vitro PHDF and HEKa proliferation

The effects of HA supplementation to tissue culture plastic (TCP) and Integra^® on PHDF and HEKa proliferation were compared to UI through MTT assays (Roche). This colorimetric assay provides accurate cell number estimation through the conversion of tetrazolium salt to formazan (Supplementary Fig. S3). Cells were synchronized in the G1/G0 of the cell cycle by incubation in 0.1 % (v/v) serum complete DMEM for 24 hours. After this period, cells were trypsinized and aliquoted for each HA concentration and added at 1 × 10⁴ cells per well. The assay was performed to the manufacturer's instructions. Briefly, the culture media was replaced by 100 μl of complete DMEM. 10 μl of MTT tetrazolium salt dye solution was added in the dark. The covered plates were incubated at 37°C for 4 hours. Insoluble formazan produced by viable cells was solubilised in 100 μl of test mix overnight at 37°C. 100 μl of the dye solution was removed and the optical density measured by a spectrophotometer at wavelength optical density (OD) 550 nm (corrected for OD 690 nm). Proliferation was assessed after 1, 2, 3, 7 and 14 days of culture. Results represent means of three independent assay reactions, for three different cell populations.

Cell cycle analysis

To investigate the influence of HA supplementation on PHDF progression through the cell cycle, flow cytometric analysis was performed on propidium iodide stained cells. Cells were again synchronized as detailed above and seeded onto Integra^® supplemented with HA concentrations as above or onto UI. Cultured cells were harvested from scaffolds by trypsinization after 24, 48, 72 and 96 hours. The collected cells were pelleted, fixed in 70% ethanol (v/v) and stained with propidium iodide solution (50 μg/mL⁻¹ PI; 0.1 mg/mL⁻¹ RNAase A; 0.05% Triton X-100). Dyed cells were analyzed by the C6 Flow Cytometer (Accuri Cytometer Ltd., UK). The percentage of cells in the different stages of the cell cycle was quantified using CFlow^® Plus Analysis software (Accuri Cytometers Ltd., UK). Results represent means of three independent assay reactions pooled together, for three different PHDF populations.

Cell viability

LDH (lactate dehydrogenase) assays (Roche, UK) were used to assess whether HA supplementation influenced cell viability. LDH is released into the cell-culture supernatant when the plasma membrane is damaged. The activity of this enzyme outside of the cell to reduce the tetrazolium salt to formazan is measured and therefore a lower assay absorbance indicates higher cell viability. The assay was performed to the manufacturer's instructions. Briefly, cell culture media was removed from fibroblasts cultured on Integra^® supplemented with HA, UI and TCP. 100 μl of the removed media was aliquoted into new 96-well plates. 100 μl tetrazolium test mix was added and the reaction incubated in the dark with gentle shaking at 20°C for 30 minutes. Absorbance was measured at OD 492 nm (corrected for OD 690 nm). Results represent means of three independent assay reactions, for three different PHDF populations. To assess the viability of ex vivo tissue during the culture period colorimetric LDH assays were also performed through the removal of culture media as above. Results represent means of three independent assay reactions, for three ex vivo cultures.

Analysis of in vitro cell invasion of Integra^®

PHDF movement vertically into, and horizontally across, Integra^® was investigated. The effect of HA supplementation on vertical migration into Integra^® was assessed through hematoxylin and eosin (H&E) staining of paraffin-embedded cross-sections.

Migration horizontally across Integra^® was assessed using CytoSelect™ 24-Well Wound Healing Assay (Cell Biolabs, Inc., USA). Fibroblasts were synchronized and subsequently seeded around a stopper area and left to attach for 24 hours. Cell stoppers were removed and scaffolds gently washed with DMEM to remove unattached cells. Wash medium was replaced with complete DMEM and cells were cultured for 48 and 72 hours. Scaffolds were formalin fixed and washed in PBS. Attached cells were permeabilized in 0.1% (v/v) Triton-X solution, washed with PBS and their nuclei stained with DAPI. Cells that had migrated into the wound area were counted across three wound fields for three PHDF populations and the mean calculated.

Cell signaling activation

HA-induced activation of signaling molecules in PHDFs was assessed by a proteome profiler array (R&D Systems, USA). Cells were cultured on UI or Integra^® supplemented with 1.5 mg/mL-¹ HA for 3 and 7 days. Cells were lysed and total protein quantified through a Non–Interfering™ Protein Assay (Merck Biosciences, UK). Protein was normalized to 200 μg in a final volume of array buffer. Negative controls consisted of PBS. The assay was performed according to the manufacturer's instructions (see Supplementary Methods). The pixel density of each capture antibody spot was analysed through ImageJ software (NIH Image). The average pixel density for each duplicate was adjusted by background subtraction and the relative change in protein phosphorylation compared between HA supplemented Integra^® and UI.

Histology of artificial wound models

Ex vivo wound models were removed from culture at 7, 14 and 21 days for analysis of wound healing through histochemistry of wound cross-sections. Tissue was fixed in 4% (v/v) formalin and processed. Wounds were bisected through the center of the wound and placed en face for paraffin embedding. For wound closure measurement, consecutive cross-sections were taken through the center of the wound (25 μm apart). Sections were routinely de-paraffinized, stained for nuclei with hematoxylin, counterstained with eosin and mounted for microscopic observation.

Statistical Analysis

For comparisons of cell proliferation, cytotoxicity, cell migration and protein phosphorylation, results were expressed as means ± SD. Post-hoc Dunnett's 2-sided t-tests were used to compare the UI (control group) against all HA supplemented Integra^® scaffolds individually at each time point. Data for protein phosphorylation assay was statistically tested using independent t-tests. All data were analyzed using IBM SPSS Statistics Software version 19.0 (SPSS Inc., Chicago, IL, USA). Significant values were considered as those where p<0.05.

Effect of HA supplementation on PHDF and HEKa in vitro proliferation

When HA was supplemented to PHDFs grown on TCP substrates, a significant increase was observed in cell proliferation over two weeks at all test concentrations in comparison to unsupplemented TCP (Fig. 1A). When HA was supplemented to Integra^® a comparable effect was observed. Initial attachment and proliferation between seeding and 1 day culture was enhanced in 0.15 mg/mL⁻¹ HA (p<0.001), 1 mg/mL⁻¹ HA (p≤ 0.05) and 1.5 mg/mL⁻¹ HA (p<0.001) scaffolds in comparison to UI (Fig. 1B). This result suggests that the number of cells attaching to the scaffold and in particular attaching in a manner that facilitates rapid proliferation is increased with HA supplementation at concentrations up to 1.5 mg/mL⁻¹. Cell numbers continue to be elevated with HA supplementation after 7 days in culture, with 1.5 mg/mL⁻¹ HA and 2 mg/mL⁻¹ HA highly significant (p<0.001). After 14 days, proliferation in HA-supplemented scaffolds was elevated in comparison to UI, with cell density approaching a dynamic equilibrium due to confluence.

OPEN IN VIEWER

Fig. 1

Effect of HA supplementation on the in vitro proliferation and viability of PHDF and HEKa cultured on Integra^® and TCP. Assessed through MTT assays (A) Effect of HA supplementation on PHDF and HEKa proliferation on TCP (B) Effect of HA supplementation on PHDF and HEKa proliferation on Integra^®. (Optical Densities are means of three independent experimental triplicates. SD) (C) LDH assay. Effect of HA supplementation on PHDF viability on Integra^® (Optical Densities are means of three independent triplicates ± SD). (* = 0.05 >p<0.001; ** p<0.001), HA supplementation increased fibroblast viability over UI at 24 hours for 0.15 mg/mL⁻¹ (p = 0.017), at 48 hours 1 mg/mL⁻¹ (p = 0.007), 1.5 mg/mL⁻¹ (p = 0.001), at 72 hours 0.15 mg/mL⁻¹ (p = 0.09), 1 mg/mL⁻¹ (p<0.001), 1.5 mg/mL⁻¹ (p<0.001) and 2 mg/mL⁻¹ (p<0.001). Significantly higher cell viability was recorded using two dimensional tissue culture plastic, showing statistical significance at 2, 3 and 7 days (p = 0.016, p<0.001 and p<0.001, respectively) (D) Cell Cycle Analysis through flow cytometry of propidium iodide stained cells after culture on Integra^® supplemented with HA or unsupplemented Integra^®. Percentage cells in each stage of the cell cycle (three independent experimental triplicates combined, ± standard error).

HA supplementation to HEKa cultured on TCP and Integra^® also produced elevated proliferation. Keratinocytes cultured on TCP and Integra^® showed similar proliferation profiles, with a proliferation phase up to 3 days followed by stagnation in cell number (Figs. 1A and 1B). During the latter stages of culture (>7 days), differentiated cells were observed in cultures with and without HA, which is likely to be a result of high cell density (28).

LDH-based colorimetric assays (Fig. 1C) demonstrated that there was no cytotoxic effect associated with HA-supplementation. In fact, HA supplementation increased fibroblast viability over UI at 1 day for 0.15 mg/mL⁻¹, at 2 days 1 mg/mL⁻¹, 1.5 mg/mL⁻¹, at 3 days 0.15 mg/mL⁻¹,1 mg/mL⁻¹, 1.5 mg/mL⁻¹ and 2 mg/mL⁻¹. Significantly higher cell viability was recorded using two TCP, showing statistical significance at 2, 3 and 7 days (p = 0.016, p<0.001 and p<0.001, respectively). This could be expected as TCP is a rigid, 2-D surface promoting attachment and providing optimal nutrient transfer conditions.

Cell cycle analysis shows a predominance of PHDFs in G0/G1 phase, especially up to 24 hours in culture, which could be as a result of cell synchronization prior to seeding. With HA, an increased number of PHDFs are in S-phase and in some cases G2/M, particularly in the first 72 hours (Fig. 1D) and is most pronounced following supplementation with 1.5 mg/mL⁻¹ and 2 mg/mL⁻¹.

Effect of HA supplementation of Integra^® on PHDF migration

Horizontal migration into the created wound area on the surface of Integra^® was influenced by HA supplementation (Fig. 2). After 48 hours 1 mg/mL⁻¹ and 1.5 mg/mL⁻¹ HA supplementation significantly increased cell migration into the wound area in comparison to UI. At 72 hours, all HA-supplemented scaffolds showed significantly more cell migration into the wound area compared to UI. A HA-concentration-related increase in cell migration was observed in concentrations up to 1.5 mg/mL⁻¹. However, the highest test concentration of HA (2 mg/mL⁻¹) was found to have an effect similar to 0.15 mg/mL⁻¹ supplementation on migration, though still significantly higher than UI.

OPEN IN VIEWER

Fig. 2

Effect of HA supplementation on the horizontal migration of primary human dermal fibroblasts across surface of Integra^®. Cell migration into wound healing assay cell free area was assessed at 48 and 72 hours, denoted here by the area within the yellow dashed line. Cells were fixed and nuclei DAPI stained (false colored green for clarity). Images are representative images of three independent experiments. After 48 hours 1 mg/mL⁻¹ HA (p = 0.007) and 1.5 mg/mL⁻¹ HA (p = 0.001) supplementation led to significantly more cell migration into the wound area in comparison to UI. At 72 hours, all HA supplemented scaffolds showed significantly more cell migration into the wound area compared to UI: 0.15 mg/mL⁻¹ (p = 0.009), 1 mg/mL⁻¹ (p<0.001), 1.5 mg/mL⁻¹ (p<0.001) and 2 mg/mL⁻¹ (p<0.001). Cell numbers are means of a minimum of three independent experiments 0 SD (* = 0.05 >p<0.001; ** p<0.001) 48 hours (×10 magnification), 72 hours (×4 magnification).

The vertical migration of cells into the Integra^® scaffold was analyzed through staining of paraffin-embedded cross-sections. H&E staining of the 5 μm sections indicated that cell infiltration was restricted to the upper portions of Integra^® in this system (Supplementary Fig. S4). PHDFs across all substrates remained on the surface of Integra^® often forming layers of several cells.

Effect of HA supplementation on the activation of proliferation/survival MAPK pathway

To investigate the activation of cell signaling pathways Human Phospho-MAPK arrays were performed comparing the apparent most effective HA supplementation concentration (1.5 mg/mL⁻¹) and UI (Fig. 3). After 3 days in culture there are statistically significant increases in the phosphorylation status of several Erk/MAPK pathway proteins involved in growth/proliferation/cell cycle progression (Akt1/2/3, Erk1/2, RSK1/2) and survival (Akt1/2/3, HSP27, JNKs). At 7 days in culture, there remains up-regulated expression, though Akt1/2, JNK1/3/Pan and RSK1 were no longer statistically significant.

OPEN IN VIEWER

Fig. 3

Relative levels of phosphorylated Erk /MAPK signalling pathway proteins in cells cultured on UI and Integra^® Dermal Substitute supplemented with 1.5 mg/mL⁻¹ HA after 3 and 7 dDays in culture. Extracted protein was collected from scaffolds supplemented with 1.5 mg/mL⁻¹ HA and UI controls. Negative controls consisted of sterile PBS solution. Phosphorylated protein assay pixel intensity was measured and expressed as the mean of assay duplicates (protein extracted from n = 250 culture wells per substrate) SD. Graphs show relative differences in pixel density (amount of protein phosphorylation) between UI and 1.5 mg/mL⁻¹ HA supplemented Integra^®. Protein phosphorylation was up-regulated relative to unsupplemented scaffolds at 3 days and 7 days (* = 0.05>p<0.001; **p<0.001).

Effect of HA supplementation on re-epithelialization of ex vivo wound models

Scaffolds were inserted into artificial wounds and re-epithelialization measured through histochemistry up to 3 weeks in culture (Fig. 4). After 1 week, migration into the wound area was restricted to wound margins and comparable for all scaffolds. After two weeks, keratinocytes in UI models are restricted to the wound margins. Migration in 0.15 mg/mL⁻¹ HA scaffolds comparable to UI though migratory cells are more organized into tongue-like structures. In 1 mg/mL⁻¹ HA scaffolds, migratory keratinocytes close the artificial wound, forming layers several cells thick. The 1.5 mg/mL⁻¹ HA and 2 mg/mL⁻¹ HA scaffolds produced similar migration, with cells penetrating into the scaffold and secreting ECM. After three weeks, migration into wounds that were without scaffold assistance was poor. In UI, cell migration was diffuse and at wound margins epidermal stratification and ECM deposition was increased. The supplemented concentrations of HA produced keratinocyte migratory morphologies of differing cell numbers, ECM production and cell positions within scaffolds following culture. 0.15 mg/mL⁻¹ HA scaffolds promoted keratinocyte migration along scaffold surfaces, closing the wound area. At 1 mg/mL⁻¹ and 1.5 mg/mL⁻¹ HA, migration of keratinocytes became less organized and ECM deposition increased. At 2 mg/mL⁻¹ HA supplementation matrix deposition was decreased but migratory cells were observed throughout scaffolds. Cell viability was maintained, as assessed through LDH assays through the 3 weeks in culture, with HA supplementation decreasing levels of cell death, particularly up to three days (Fig. 5).

OPEN IN VIEWER

Fig. 4

Examination of wound closure through H&E stained wound cross sections after 1, 2 and 3 weeks ex vivo culture. Representative images of keratinocyte migration across Integra^® scaffolds with and without HA supplementation. Integra^®-assisted wounds promoted keratinocyte migration in comparison to unassisted wounds. HA supplementation further increased keratinocyte migration.

OPEN IN VIEWER

Fig. 5

LDH assays of cell viability in ex vivo wound models. Models showed a small, general increase in cell death up to one week and subsequent decrease. Up to three weeks in culture there was no significant increase in cell death for any scaffolds, indicating cell viability is maintained.