Serum-Free Primary Human Fibroblast and Keratinocyte Coculture

Abstract

Research has shown that the inclusion of a fibroblast cell support layer is required for the isolation and expansion of primary keratinocytes. Recent advances have provided keratinocyte culture with fibroblast-free alternatives. However, these technologies are often undefined and rely on the incorporation of purified proteins/components. To address this problem we developed a medium that used recombinant proteins to support the serum-free isolation and expansion of human dermal fibroblasts and keratinocytes. The human dermal fibroblasts were able to be isolated serum free by adding recombinant human albumin to a collagenase solution. These fibroblasts were then expanded using a serum-free medium containing recombinant proteins: epidermal growth factor, basic fibroblast growth factor, chimeric vitronectin:insulin-like growth factor-I protein, and recombinant human albumin. These fibroblasts maintained a typical morphology and expressed fibroblast markers during their serum-free isolation, expansion, and freezing. Moreover, these fibroblasts were able to support the serum-free isolation and expansion of primary keratinocytes using these recombinant proteins. Real-time polymerase chain reaction and immunofluorescence analysis confirmed that there were no differences in expression levels of p63 or keratins 1, 6, and 10 when keratinocytes were grown in either serum-supplemented or serum-free medium. Using a three-dimensional human skin equivalent model we demonstrated that these keratinocytes also maintained their ability to reform an epidermal layer. In summary, the techniques described provide a valuable alternative for culturing fibroblasts and keratinocytes using recombinant proteins.

Introduction

Keratinocytes are the most predominant cell type found in the epidermal layer of the skin and have been cultured in vitro for the production of skin grafts to treat burns,^1,2 diabetic foot ulcers,³ and oral mucosa defects.⁴ Rheinwald and Green (1975) were the first to demonstrate that these cells could be expanded in vitro using an irradiated mouse fibroblast feeder cell line 3T3 (i3T3) in medium supplemented with fetal calf serum (FCS).⁵ Although this achievement has led to several advances in clinical wound repair, there are growing concerns that this approach generally requires the use of poorly defined, animal-derived serum components and/or feeder cells. These not only are undefined components but also present a potential risk of transmitting animal-derived diseases to patients.⁶ The use of murine fibroblast feeder cells introduces another potential risk as these too can act as a potential disease source. Recent reports also indicate that human embryonic stem cells, another feeder cell-dependent cell type, have expressed the mouse fibroblast xeno-antigen Neu5Gc,^7,8 indicating the negative influences that fibroblasts can have within these culture systems.

To address these challenges, several investigators have attempted to develop serum-free culture methods for keratinocytes.^9–16 However, these methods often utilize components such as bovine^13,16 or porcine pituitary extract¹² or undefined commercially available substitutes,¹⁵ which present a risk of transmitting animal-derived diseases to the graft. Recent studies within our laboratory have led to the development of a culture system that successfully supports the ex vivo expansion of human keratinocytes in the absence of serum and serum-derived factors.^14,17 This fully defined medium consisted of vitronectin (VN), insulin-like growth factor-I (IGF-I), IGF-binding protein-5, and epidermal growth factor (EGF). However, cells grown in this medium required the presence of an i3T3 for their expansion.

In view of the apparent critical requirement for fibroblast feeder cells, several research groups have investigated the use of human-derived feeder cell layers to replace the murine 3T3 cells. For example, human dermal fibroblasts (HDFs)^9,11,18,19 and fetal lung fibroblasts (MRC-5 cells)⁹ have both been investigated as an alternative to murine 3T3 cells. However, isolation of both fibroblasts and keratinocytes in these studies required the use of FCS, reintroducing a potential point of contamination. Thus, although these techniques remove the need for xeno-derived feeder cell layers, the keratinocytes require serum-containing medium and allogeneic feeder cells for their isolation and expansion. As such, it is critical that fibroblasts are isolated and expanded in serum-free conditions if they are to be used for the culture of human keratinocytes for use in clinical applications.

With this in mind we have investigated the potential of using various combinations of recombinant proteins to support the isolation and expansion of HDF. Specifically, combinations of recombinant human albumin (HA), a VN:IGF-I chimeric protein,²⁰ EGF, and basic fibroblast growth factor (bFGF) were investigated. The ability of these fibroblasts to support the serum-free isolation and expansion of keratinocytes was then investigated using the same combination of proteins. These were investigated because IGF-I, a key component in the recombinant chimeric protein VN:IGF-I, has previously been shown to increase keratinocyte proliferation,^21,22 while VN, the other component of the recombinant chimeric protein, has been demonstrated to enhance the effects of IGF-I.^14,23 In addition, EGF is commonly added to keratinocyte cultures as it has been shown to delay senescence of keratinocytes,²⁴ while bFGF has also been shown to be important for keratinocyte migration,^25–27 as well as acting as a mitogen for keratinocytes.^26,28 It was therefore feasible that a combination of HA, chimeric VN:IGF-I, bFGF, and EGF might support the growth of keratinocytes in the absence of any xeno-derived components. The specific aims of this study were, first, to investigate whether these proteins could support the isolation and expansion of primary HDF in the absence of serum. Second, we aimed to investigate if these HDFs could support the serum-free isolation and expansion of primary keratinocytes.

Materials and Methods

Skin samples and cell lines

The keratinocyte and fibroblast cells used in these experiments were obtained with consent from patients undergoing breast reductions or abdominoplasties. Ethics approval to use this tissue was obtained from the Queensland University of Technology Human Research Ethics Committee (ID: 3673H), as well as from the Royal Children's (no. 2007/021), St. Andrews and Wesley Hospitals, Brisbane, Australia. The human foreskin fibroblast (HFF) cell lines CCD-1137SK and HFF-1 (ATCC, Manassas, VA), as well as the murine fibroblast 3T3 cell line (ATCC) were also used.

Culture of 3T3 cells

3T3 cells were expanded in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Mount Waverley, Victoria, Australia) supplemented with 5% FCS (Hyclone, South Logan, UT), 2 mM l-glutamine (Invitrogen), and 1000 IU/mL penicillin/streptomycin (Invitrogen) in 5% CO₂ at 37°C. These cells were mitotically inactivated by gamma-irradiation in suspension (two doses of 25 Gy; Australian Red Cross Blood Services, Brisbane, Queensland, Australia) and seeded into flasks or culture dishes (Thermo Fisher Scientific, Rochester, NY) at a density of 2.67 × 10⁴ cells/cm² for use as a feeder cell layer for primary keratinocytes.

Culture of HFF cell lines

The HFF cell lines were established by seeding 5.5–6.5 × 10⁶ HFF-1 cells or 3.2 × 10⁶ CCD-1137Sk cells into a 75 cm² culture flask (Thermo Fisher Scientific) in medium containing 10% FCS at 37°C in 5% CO₂. The cells were subcultured into serum-free conditions after three passages (P16 for HFF-1 and P5 for CCD-1137Sk). Briefly, 2 × 10⁵ cells were seeded into 25 cm² culture flasks (Thermo Fisher Scientific) that had been precoated with recombinant human type 1 collagen (5 μg/cm²; FibroGen, South San Francisco, CA) for 2 h. These cells were then grown in our serum-free basal medium, which we termed “Stripped Green's Plus” (SG+) medium. This medium contained DMEM (Invitrogen) and HAMS-F12 (Invitrogen) in a 3:1 ratio supplemented with 2 mM l-glutamine (Invitrogen), 1000 IU/mL penicillin/streptomycin (Invitrogen), 0.1 mM nonessential amino acids (Invitrogen), 0.2 μM triiodothyronine (Sigma-Aldrich, Castle Hill, New South Wales, Australia), 180 μM adenine (Sigma-Aldrich), 0.1 μg/mL cholera toxin (Sigma-Aldrich), 0.4 μg/mL hydrocortisone (Sigma-Aldrich), 5 μg/mL transferrin (Sigma-Aldrich), 1:1000 dilution (v/v) lipid concentrates (Invitrogen), and 0.04 × trace elements B and C (Mediatech, Manassas, Virginia). SG+ medium was then supplemented with 50 ng/mL recombinant human bFGF (Millipore, Sydney, New South Wales, Australia), 50 ng/mL recombinant human EGF (Invitrogen), 0.5 μg/mL recombinant human chimeric VN:IGF-I (Tissue Therapies, Brisbane, Queensland, Australia), and 2 mg/mL recombinant HA (Albucult^®; Novozymes Biopharma UK, Nottingham, United Kingdom) (SG + FIB). After the cells had been grown in serum-free medium for at least two passages, different concentrations of HA (1–2 mg/mL) were compared. These cells were then mitotically inactivated by gamma-irradiation in suspension (two doses of 25 Gy; Red Cross Blood Services) and 5 × 10³ cells/cm² were then seeded into culture dishes to be used as a feeder cell layer for primary keratinocytes.

Primary fibroblast cell culture

Primary fibroblasts were isolated from skin by mincing the dermal region of the skin into small pieces, followed by digestion with 0.05% collagenase A (Invitrogen) in DMEM with 5–15 mg/mL HA overnight at 37°C with 5% CO₂. The slurry was then centrifuged at 425 relative centrifugal force (RCF) for 10 min and resuspended in SG+ medium. These cells were then seeded into flasks (2 × 10⁶ cells/T75 flask; Thermo Fisher Scientific) that had been precoated for 2 h with recombinant human type 1 collagen (5 μg/cm²) or in uncoated flasks.

Primary fibroblasts were grown in SG+ medium supplemented with 50 ng/mL bFGF, 50 ng/mL EGF, 0.5 μg/mL chimeric VN:IGF-I, and 1 mg/mL HA (SG+FIB). As a control, fibroblasts were also grown in a medium containing 5% FCS. These primary fibroblasts were grown at 37°C in 5% CO₂ and the medium was replenished every 2 days. Once the cells reached confluence, they were passaged by gentle washing with phosphate-buffered saline minus calcium and magnesium (PBS−; Invitrogen), followed by detachment using TrypLE (Invitrogen). The cell suspension was then pelleted at 106 RCF for 5 min and resuspended in medium. The fibroblasts were reseeded into flasks precoated with recombinant laminin (5 μg/cm²; Millipore) or recombinant human type 1 collagen (5 μg/cm²; FibroGen), or into uncoated flasks. In addition, fibroblasts initially grown in serum were passaged into serum-free medium after at least two initial passages in serum-supplemented medium to determine if SG+FIB medium could support fibroblasts that had previously been exposed to serum. Once the fibroblasts were expanded, they were mitotically inactivated by gamma irradiation (two doses of 25 Gy; Red Cross Blood Services) and 5 × 10³ cells/cm² were seeded into culture dishes precoated with recombinant human type 1 collagen (5 μg/cm²) or into uncoated dishes for use as a feeder cell layer for primary keratinocytes. This fibroblast extraction and culture was performed on skin from three separate patients and each lot was kept separate to give n = 3 fibroblast feeder layers for each keratinocyte coculture experiment.

Freezing of human fibroblasts

Human fibroblasts that had been expanded in serum were frozen in liquid nitrogen by suspending 1 × 10⁶ cells/mL in 90% FCS and 10% dimethyl sulfoxide. For serum-free cultures, 1 × 10⁶ cells/mL were suspended in CryoStor™ CS-10 (In Vitro Technologies, Melbourne, Victoria, Australia) and frozen in liquid nitrogen. The viability of these cells following storage for 3 weeks was assessed by cell counting and serial expansion.

Development of serum-free culture conditions for primary keratinocytes

Primary keratinocytes were isolated from skin samples using a modification of the method described by Rheinwald and Green.⁵ Briefly, the epidermis was separated from the dermis by digesting the skin in 0.125% trypsin (Invitrogen) in PBS− at 4°C overnight. Keratinocytes were then isolated by gentle scraping at the epidermal–dermal junction and were pelleted by centrifugation at 106 RCF for 5 min. The dermal region of the skin samples was saved for isolation of fibroblasts. Keratinocytes were seeded at a density of 4 × 10⁴ cells/cm² into 10 cm² culture dishes (Thermo Fisher Scientific) previously seeded with a monolayer of i3T3 cells or human fibroblasts. The keratinocyte extraction was performed on skin from three separate patients and each lot was kept separate to give n = 3 keratinocytes to seed onto the previously described human fibroblast feeder layers or 3T3 cells.

SG+ medium was supplemented with chimeric VN:IGF-I (0.5–1 μg/mL), bFGF (20–100 ng/mL), EGF (20–100 ng/mL), and recombinant HA (0–4 mg/mL) (SG+GF). Once the keratinocytes reached confluence, the feeder cells were removed using 0.02% ethylenediaminetetraacetic acid (EDTA)/PBS− (10 min/37°C) and the keratinocytes were passaged using TrypLE. Subcultures were established using the same density of feeder cells, but this time the keratinocytes were seeded at only 4 × 10³ cells/cm². For subcultures the culture dishes were not precoated with recombinant collagen. Keratinocytes were grown in Green's medium containing DMEM (Invitrogen) and HAMS-F12 (Invitrogen) in a 3:1 ratio, supplemented with 2 mM l-glutamine (Invitrogen), 1000 IU/mL penicillin/streptomycin (Invitrogen), 0.1 mM nonessential amino acids (Invitrogen), 0.2 μM triiodothyronine (Sigma-Aldrich), 180 μM adenine (Sigma-Aldrich), 0.1 μg/mL cholera toxin (Sigma-Aldrich), 0.4 μg/mL hydrocortisone (Sigma-Aldrich), 5 μg/mL transferrin (Sigma-Aldrich), 10% FCS (Hyclone), 10 ng/mL EGF (Invitrogen), and 1 μg/mL insulin (Sigma-Aldrich) as a positive control.

MTT and WST-1 assays

The 3-(4,5-dimethylthiazol-2-yl)-2-5-diphenyltetrazolium bromide (MTT) and 4-[3-(4-iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate (WST-1) assays were used to assess differences in metabolic activity of keratinocytes and fibroblasts, respectively. The MTT assay was initially used for both cell types as it is already well established within our laboratory. However, after reviewing the literature we discovered that the WST-1 assay was more stable and less disruptive to the cells. In view of this, for the later experiments using fibroblast the WST-1 assay was employed.

For the MTT assay, keratinocytes were grown to confluence on a monolayer of i3T3 cells in Green's medium. These cells were then serum starved for at least 4 h and the feeder cells were removed using 0.02% EDTA in PBS−. The keratinocytes were then seeded into 1.8 cm² culture plates (Thermo Fisher Scientific) at a density of 5 × 10⁴ cells per well and cultured for 72 h. Keratinocytes grown in Green's medium with i3T3 feeder cells were used as a positive control. Different concentrations of EGF (10–100 ng/mL), bFGF (10–100 ng/mL), and chimeric VN:IGF-I (0.25–1 μg/mL), as well as combinations of EGF ± bFGF ± chimeric VN:IGF-I were investigated for their effect on the metabolic activity of keratinocytes. These growth factors were added to SG+ medium supplemented with 2 mg/mL HA. SG+ medium containing only HA was utilized as a negative control. Following cultivation, the cells were washed with PBS− and the feeder cells (i3T3s) were removed by adding 0.02% EDTA in PBS− where necessary. The cultures were then incubated in 0.5 mg/mL MTT reagent (Sigma-Aldrich) for 1 h at 37°C and the converted formazan product was eluted using 300 μL of dimethyl sulfoxide (Sigma-Aldrich). This formazan product (100 μL) was then transferred into a 96-well plate and the absorbance (540–630 nm) was measured.

The WST-1 assay was used to assess differences in metabolic activity between fibroblasts grown in different concentrations of bFGF (10–100 ng/mL), EGF (10–100 ng/mL), and chimeric VN:IGF-I (0.25–1 μg/mL), as well as combinations of these proteins in SG+ medium supplemented with 2 mg/mL HA. For the positive and negative controls, the fibroblasts were also grown in serum-supplemented conditions and in SG+ medium, respectively. The fibroblasts were seeded into 96-well (0.32 cm²) plates (Thermo Fisher Scientific) at a density of 5 × 10³ cells per well and grown for 24 h using SG+ medium supplemented with the same combinations of proteins described above for the keratinocytes. WST-1 reagent (10 μL; Roche, Castle Hill, New South Wales, Australia) was added to each well (1:10 dilution, v/v) and incubated for 3 h at 37°C, 5% CO₂. The plates were subsequently shaken on a plate shaker and the absorbance (440–620 nm) was measured.

To monitor cell proliferation, keratinocyte and fibroblast expansion cultures were counted by trypan blue staining after passage 1, 2, 3, and 4. Cell counts of keratinocytes cultured on i3T3 or irradiated HDF feeder cells in either Green's or SG+GF medium and cell counts of the HDFs cultured in serum-containing or SG+GF medium were performed. The cell counts were performed on five patients for the keratinocytes and four patients for the fibroblasts and then averaged.

Immunofluorescence

Immunohistochemistry of cultures was performed to determine if keratinocytes maintained markers indicative of a proliferative state. This was performed using monoclonal antibodies toward keratin 1/10 (1:200 dilution; US Biological, Swampscott, MA), p63 (1:100 dilution; Neomarkers, Fremont, CA), and keratin 6 (undiluted; Fitzgerald Ind., Concord, MA). Keratins 1 and 10 are markers expressed by differentiated keratinocytes, whereas p63 is a marker of basal keratinocytes and keratin 6 is expressed by hyperproliferative keratinocytes. In addition, immunofluorescence was undertaken on fibroblasts cultivated in serum-free conditions to confirm that the cells being expanded were in fact fibroblasts. This was assessed by probing the cells with an antifibroblast antibody (1:50 dilution; Millipore) and a monoclonal antibody directed toward Thy-1, a cell surface marker expressed by fibroblasts (1:50 dilution; BD Biosciences, San Jose, CA).

Briefly, cultured cells were fixed using 4% paraformaldehyde (Sigma-Aldrich; 25°C/15 min) and stored in PBS−. Cells were then permeabilized in 0.1% Triton X-100 (Merck, Darmstadt, Germany) in PBS− for 10 min at room temperature and blocked using 2% bovine serum albumin (Sigma-Aldrich) in PBS− (blocking buffer) for 30 min. The primary antibody was then added and the cells were left to incubate for 1 h at room temperature. The cells were then washed three times for 5 min per wash using blocking buffer. Alexa Fluor 488 goat anti-mouse immunoglobulin G (IgG) secondary antibody (1:200 dilution; Invitrogen) was added and left for 1 h in the dark. The cells were then washed three times before mounting on slides using an antifade mounting medium that contains 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA). Imaging was performed using the Nikon Eclipse TE2000-U fluorescent microscope (Nikon, Lidcombe, New South Wales, Australia). For both the fibroblasts and the keratinocytes, the negative control involved treatments in which the primary antibody was omitted. A mouse IgG₁ isotype control (1:500 dilution; R&D Systems, Gymea, New South Wales, Australia) was included as an additional negative control for the fibroblast immunofluorescence experiments.

Real-time polymerase chain reaction

Real-time polymerase chain reaction (PCR) was used to assess the expression of genes that are markers for more differentiated (keratins 1 and 10) and less differentiated (keratin 6 and p63) keratinocytes. To perform this reaction, total RNA was extracted from the cells using the Qiagen RNeasy plus mini kit (Qiagen, Doncaster Victoria, Australia) as per the manufacturer's instructions. cDNA was then synthesized from 2 μg of total RNA using SuperScript III Reverse Transcriptase (Invitrogen) and random primers (Invitrogen). Briefly, random primers were added to 2 μg of total RNA and allowed to anneal by heating for 5 min at 65°C. The samples were then held on ice for 2 min before adding the SuperScript III Reverse Transcriptase and buffers. cDNA was formed by heating these samples for 5 min at 25°C, followed by 1 h incubation at 50°C and then 15 min at 70°C.

Real-time PCR was performed on 100 ng of cDNA template using the SYBR Green PCR Master Mix (Qiagen) and the primers (Sigma-Aldrich) described in Table 1. These primers were designed using the Invitrogen Oligoperfect designer and span an exon/intron boundary. PCR reactions were performed in triplicate with an initial 10 min activation step at 95°C, followed by 45 cycles of 95°C for 20 s, 55°C for 10 s, 60°C for 30 s, and 72°C for 40 s. Relative expression units were calculated by normalizing the difference in cycle threshold value (ΔCt) values of the genes to the ΔCt of glyceraldehyde 3-phosphate dehydrogenase. A “no cDNA template” was added as a control for contamination.

Table 1.

Primer Sequences Used for Real-Time Polymerase Chain Reaction

Gene	Description	Forward primer sequence	Reverse primer sequence
P63	Basal cell marker	5′-AGCAGCAAGTTTCGGACAGT	5′-CATCATCTGGGGATCTTCGT
Keratin 6	Proliferation marker	5′-GACCTGGTGGAGGACTTCAA	5′-GTAGGCAGCATCCACATCCT
Keratin 1	Differentiation marker	5′-TGACCCTGAGATCCAAAAGG	5′-TCCAGGAACCTCACCTTGTC
Keratin 10	Differentiation marker	5′-ATGAGCTGACCCTGACCAAG	5′-TCATTTCCTCCTCGTGGTTC

Human skin equivalent experiments

Three-dimensional human skin equivalent (HSE) models were prepared by initially digesting skin in a 1 M NaCl solution overnight at 37°C to separate the epidermis from the dermis. The epidermis was then peeled away from the dermis to produce a decellularized, deepidermized dermis (DED). The DED was then washed at least five times by submerging it in 1000 IU/mL penicillin/streptomycin in DMEM for at least 2 h. The DED was then cut into 1 cm² pieces and submerged in either Green's or SG+ medium for 2 h. Sterile stainless-steel rings with a silicon washer base (0.7 mm internal diameter; Aix Scientifics, Aachen, Germany) were then placed on the papillary side of the DED and 600 μL of Green's medium or SG+ medium with 100 ng/mL bFGF, 100 ng/mL EGF, 0.5 μg/mL chimeric VN:IGF-I, and 2 mg/mL HA (SG+GF) was added to the outside of the rings. Keratinocytes (2 × 10⁴) were then seeded inside the rings in 200 μL of growth medium. Growth medium without keratinocytes was also utilized as a negative control. After 72 h incubation at 37°C in 5% CO₂ the rings were removed and the HSE models were transferred to stainless-steel grids. Growth medium was then added until it was in level with the DED but not over the top, thus creating an air–liquid interface. After 7 days at the air–liquid interface the HSE models were removed and fixed in formalin for at least 24 h before being processed by paraffin sectioning. Each HSE model was cut into 3 μm sections and stained using hematoxylin and eosin. Immunohistochemical staining of the reconstituted epidermis was performed as described previously.¹⁴ Briefly, immunoperoxidase staining for p63, keratin 1/10/11, and keratin 14 was performed using a MACH 4 Universal HRP-polymer kit with DAB (Biocare, Concord, CA). Primary antibodies to p63 and keratin 14 were used at 1:100, and the keratin primary antibody was used at 1:200. Negative control involved treatments in which the primary antibody was omitted. Staining was performed on the reconstituted epidermis from three different patients.

Statistics

Data acquired from the MTT and WST-1 assays were analyzed using a linear mixed model equation. This test was used as it accounts for differences attributed to patient variability. Data acquired from real-time PCR analysis were analyzed using Tukey's test on unpooled samples. Significance for both tests was determined at the 0.05 level. All assays were performed in triplicate and repeated using skin from at least three different patients.

Results

Metabolic response of fibroblast cultures to bFGF, EGF, and chimeric VN:IGF-I

The WST-1 assay was used to determine which concentrations of growth factors would provide the optimal growth conditions for the isolation and expansion of primary human fibroblasts. Metabolic responses of fibroblasts after 24 h exposure toward different concentrations of EGF, bFGF, and chimeric VN:IGF-I were investigated (Fig. 1). A significant response (p < 0.05) compared with our serum-free Green's medium (SG+ medium) with no added growth factors was observed when 0.5 μg/mL VN:IGF-I was added. It was demonstrated that this level of metabolic activity did not significantly vary when 0.5 μg/mL VN:IGF-I was added in combination with either bFGF or EGF. However, when bFGF and EGF were added in combination with 0.5 μg/mL VN:IGF-I, an increase in metabolic activity was demonstrated, which was significantly higher than that observed when VN:IGF-I was added alone or with only bFGF or EGF added. All of the serum-free culture conditions that were investigated were found to be significantly lower than the serum-supplemented control.

FIG. 1.

Metabolic activity of human fibroblasts cultured in different conditions. Human fibroblasts were grown for 24 h in medium containing 5% FCS or in SG+ medium supplemented with 2 mg/mL HSA and growth factors as indicated. All treatments, with the exception of the FCS and SG+ controls, also contained 0.5 μg/mL VN:IGF-I. Stars (*) indicate a significant difference when compared with SG+ without chimeric VN:IGF-I (negative control), circles (•) indicate a significant difference when compared with SG+ with chimeric VN:IGF-I, the symbol # indicates metabolic responses that are statistically different when compared with corresponding treatment without bFGF, and triangles (Δ) indicate metabolic responses that are statistically different when compared with corresponding treatment without EGF. All treatments were statistically different to 5% FCS. Significance determined using a linear mixed model (p < 0.05). Bars represent mean ± SEM (n = 4 patients). HSA, human serum albumin; FCS, fetal calf serum; SG+, Stripped Green's Plus; VN, vitronectin; IGF-I, insulin-like growth factor-I; bFGF, basic fibroblast growth factor; EGF, epidermal growth factor; SEM, standard error of the mean.

Serum-free isolation and expansion of HDFs

Isolation of HDFs in the absence of serum was achieved by digesting the dermal region of the skin in 0.05% collagenase A in DMEM with 5, 10, or 15 mg/mL of HA. Fibroblasts isolated under these conditions were labeled Fib123, Fib126, and Fib128, respectively. It was found that fibroblasts isolated using 10 or 15 mg/mL of HA had a higher rate of proliferation and became confluent in a shorter period of time (12 days) than those isolated with 5 mg/mL HA (23 days) (Supplemental Table S1, available online at www.liebertonline.com/ten). However, all of these fibroblasts required a longer period of time to reach confluence compared with those isolated in serum (6 days; Supplemental Table S1). In addition, fibroblasts isolated with 10 or 15 mg/mL HA required a recombinant collagen type I-coated tissue culture surface (5 μg/cm²) for their initial expansion, whereas those isolated with only 5 mg/mL of HA proliferated more rapidly on uncoated plates. In addition to these experiments examining dermal fibroblasts, HFF-3 were also able to be isolated in serum-free conditions using 0.05% collagenase A in DMEM with 5 mg/mL HA. These HFFs were initially expanded on recombinant laminin-coated plates (5 μg/cm²) before being passaged onto collagen type I-coated plates (data not shown).

On the basis of the results from the WST-1 assay, a combination of 50 ng/mL bFGF, 50 ng/mL EGF, 0.5 μg/mL chimeric VN:IGF-I, and 2 mg/mL HA was added to the SG+ medium. This combination was demonstrated to support the serum-free expansion of primary fibroblasts and HFF cell lines. These conditions were further optimized to reduce the amount of HA required (1 mg/mL) for serial propagation. In addition, it was found that fibroblasts exhibited little difference in morphology and growth rates when grown on collagen type I-coated (5 μg/cm²) or uncoated flasks (data not shown). Probing the fibroblasts for Thy-1 expression, as well as with an antifibroblast antibody, demonstrated that the fibroblasts isolated in serum-free medium had similar immunoreactivity to those isolated using serum-supplemented medium (Fig. 2).

FIG. 2.

Morphology of fibroblasts isolated and grown in serum-free conditions or in serum-supplemented conditions. Dermal fibroblasts (Fib123) maintained a similar morphology when grown in serum-supplemented media (a) or serum-free media (e). Immunofluorescence micrographs demonstrated similar expression of Thy-1 when grown under serum-supplemented conditions (b) or in serum-free media (f). Fibroblasts exhibited immunoreactivity when probed with an antifibroblast antibody when grown under either serum-supplemented (c) or serum-free (g) conditions. Minimal immunofluorescence was observed when the primary antibodies were omitted (d) or when a mouse immunoglobulin G isotype control was used (h). Both scale bars represent 100 μm. Color images available online at www.liebertonline.com/ten.

In addition to isolating and expanding fibroblasts in the absence of serum, we were also able to freeze and store them in liquid nitrogen using the commercially available serum-free freezing solution CryoStor (In Vitro Technologies). These fibroblasts were shown to maintain a high level of viability upon thawing (>90%) based on their ability to exclude trypan blue stain and were then also able to be serially passaged in serum-free medium (data not shown). In addition, probing these fibroblasts for expression of Thy-1, as well as with the antifibroblast antibody, revealed similar immunoreactivity compared with fibroblasts grown in serum (data not shown).

Metabolic responses of primary keratinocytes to bFGF, EGF, and chimeric VN:IGF-I

The effects of bFGF, EGF, and chimeric VN:IGF-I on the metabolic activity of primary keratinocytes was also investigated using MTT assay (Fig. 3). Serum-supplemented culture medium (Green's medium) with i3T3 feeder cells and SG+ medium that contained no growth factors were utilized as positive and negative controls, respectively. It was found that keratinocytes grown for 72 h in SG+ medium supplemented with 100 ng/mL bFGF, 100 ng/mL EGF, 1 μg/mL chimeric VN:IGF-I, and 2 mg/mL HA with and without a feeder layer of i3T3 cells exhibited an increase in metabolic activity that was statistically different (p < 0.05) to the negative control. This level of metabolic activity was demonstrated to be significantly higher (p < 0.05) than when VN:IGF-I was absent. In addition, these cells demonstrated no significant difference in metabolic activity compared with those cultured using Green's medium with i3T3 feeder cells.

FIG. 3.

Analysis of metabolic responses of primary keratinocytes cultured in different growth conditions. Primary keratinocytes were grown for 72 h in SG+ medium supplemented with 2 mg/mL HSA and varying concentrations of bFGF, EGF, and chimeric VN:IGF-I. Keratinocytes were also grown in Green's medium with i3T3 cells and in SG+ medium as positive and negative controls, respectively. Stars (*) indicate a significant difference in metabolic activity when compared with SG+ without any chimeric VN:IGF-I added (negative control). Circles (•) indicate a significant difference in metabolic activity when compared with SG+ with 1 μg/mL VN:IGF-I. The symbol # indicates a significant difference in metabolic activity compared with SG+ with 100 ng/mL bFGF and 100 ng/mL EGF. Triangles (Δ) indicate culture conditions that exhibited a significant decrease in metabolic activity compared with the positive control (Green's medium + i3T3). Significance was determined using a linear mixed model (p < 0.05). Error bars represent mean ± SEM. n = 3 patients. i3T3, irradiated mouse fibroblast feeder cell line.

The metabolic response of keratinocytes toward different concentrations of chimeric VN:IGF-I was also investigated. The keratinocytes demonstrated a dose-dependent increase in metabolic activity when VN:IGF-I was added. It was found that the addition of 1 μg/mL VN:IGF-I was required to cause an increase in metabolic activity that was significantly higher (p < 0.05) than the negative control. This level of metabolic activity was also demonstrated to not vary significantly from the i3T3- and serum-containing (Green's medium) positive control.

Serum-free expansion of primary keratinocytes on human fibroblasts

Human fibroblasts that had been isolated and expanded in serum-free conditions were examined for their ability to support the serum-free expansion of primary keratinocytes. It was found that human fibroblasts that had been mitotically inactivated by gamma irradiation (two doses of 25 Gy) and seeded at 5 × 10³ cells/cm² were able to support primary keratinocytes for at least three passages when grown in either Green's medium or SG+ medium supplemented with 100 ng/mL bFGF, 100 ng/mL EGF, 0.5 μg/mL chimeric VN:IGF-I, and 2 mg/mL HA (SG+GF) (Fig. 4). Keratinocytes grown on human-derived feeder cells in SG+GF medium tended to become confluent in shorter periods of time (average of 5 days) compared with those grown using Green's medium with i3T3 feeder cells (average of 9 days) (Supplemental Table S1, available online at www.liebertonline.com/ten).

FIG. 4.

Passage 1 primary keratinocytes cultured in Green's medium and SG+GF medium. Keratinocytes that were grown on i3T3 cells in Green's medium (a) tended to be larger compared with those grown SG+GF medium (d). Irradiated dermal fibroblasts that had been initially expanded in serum-supplemented medium and then serially passaged using SG+FIB were shown to support the isolation and serial expansion of keratinocytes in either Green's medium (b) or SG+GF medium (e). In addition, it was demonstrated that fibroblasts that had been isolated and expanded in serum-free conditions could also support keratinocytes in Green's medium (c) and SG+GF medium (f). Arrows indicate representative cells. Scale bar represents 100 μm. GF, growth factors (keratinocytes).

Expression of markers by primary keratinocytes expanded in different conditions

To confirm that the keratinocytes grown on human-derived fibroblasts in SG+GF medium were maintaining an undifferentiated, highly proliferative state, immunostaining for several markers was performed (Fig. 5). Keratinocytes grown on i3T3 cells in SG+GF medium maintained a high level of immunoreactivity for keratin 6 (proliferation marker) and p63 (basal cell marker), whereas only minimal levels of immunoreactivity were observed for keratins 1 and 10 (differentiation markers). When these keratinocytes were grown on irradiated human fibroblast feeder cells in the SG+GF medium, high levels of keratin 6 and p63 immunoreactivity, but little to no immunoreactivity of keratins 1 and 10 were observed. Interestingly, keratinocytes grown on human fibroblast feeder cells in Green's medium exhibited greater immunoreactivity toward keratins 1 and 10 compared with the keratinocytes grown in SG+GF medium or on i3T3 cells. Further, the increased cell size observed in Figure 5g–i may be an indication that these cells are no longer in a highly proliferative state.

FIG. 5.

Immunostaining of passage 2 cultures of keratinocytes (Kc125) expanded on i3T3 cells or human fibroblast feeder cells in different media. Keratinocytes demonstrated immunoreactivity toward the hyperproliferative marker keratin 6 when grown on i3T3 in Green's medium (a), i3T3 in SG+GF medium (d), human fibroblast feeder cells in Green's medium (g), and human fibroblast feeder cells in SG+GF medium (j). In addition, keratinocytes maintained immunoreactivity toward the basal cell marker p63 when grown on i3T3 in Green's medium (b), i3T3 in SG+GF medium (e), human fibroblast feeder cells in Green's medium (h), and human fibroblast feeder cells in SG+GF medium (k). Immunoreactivity toward the differentiation markers keratins 1 and 10 was demonstrated for keratinocytes grown on human fibroblasts in Green's medium (i) or on i3T3s in SG+GF medium (f). Minimal levels of immunoreactivity were detected for keratins 1 or 10 when keratinocytes were grown on i3T3s in Green's medium (c) or on human fibroblasts in SG+GF medium (l). No fluorescence was observed when the primary antibody was omitted (inset in a). Arrows indicate representative cells. Scale bar represents 100 μm. Color images available online at www.liebertonline.com/ten.

Real-time PCR analysis of expression of genes in keratinocytes expanded in different conditions

To confirm the results observed using immunofluorescence at the mRNA level, real-time PCR was performed on keratinocytes isolated from skin from three different patients. Analysis of the p63 transcripts found that that there was no significant difference (p < 0.05) in expression between keratinocytes grown in SG+GF and Green's media (Fig. 6a). Further, no significant changes (p < 0.05) in p63 or keratin 6 expressions were noted when the keratinocytes were grown on either irradiated dermal fibroblasts or i3T3 (Fig. 6a, b). However, additionally, no significant differences in keratin 6 expression were observed between these two media, with the exception of a keratin 6 mRNA expression, which was shown to be upregulated (p < 0.05) when keratinocytes were grown on dermal fibroblasts in SG+GF medium compared with those grown in Green's medium (Fig. 6b). No significant difference (p < 0.05) in keratin 6 expression was observed when keratinocytes were grown on i3T3 in SG+GF or Green's medium.

FIG. 6.

Real-time polymerase chain reaction analysis of expression of genes in passage 2 cultures of keratinocytes (Kc124). To determine if keratinocytes grown in our serum-free conditions maintained an undifferentiated phenotype, real-time polymerase chain reaction analysis of several markers was undertaken. Keratinocytes grown in either Green's medium (dark bars) or SG+GF medium (patterned bar) were analyzed for expression of p63 (a), keratin 6 (b), keratin 1 (c), and keratin 10 (d). Relative expression units are normalized to glyceraldehyde 3-phosphate dehydrogenase expression and are expressed as the mean ± SEM. *Significant difference in expression levels when comparing to keratinocytes grown on i3T3 in Green's medium. **Expression levels that are statistically different compared with keratinocytes grown on i3T3 in SG+GF medium. ***Significant difference in expression levels when comparing keratinocytes grown on the same feeder cells in SG+GF medium to Green's medium. Significance determined using Tukey's test (p < 0.05). Bars represent mean ± SEM (n = 3).

Real-time analysis of the differentiation markers keratin 1 and 10 was also performed. An increase in keratin 1 expression was observed when keratinocytes were grown in SG+GF compared with those grown in Green's medium (Fig. 6c). Expression of keratin 1 was also shown to be downregulated when keratinocytes were grown on irradiated dermal fibroblasts compared with those grown on i3T3 (Fig. 6c). No significant difference (p < 0.05) in keratin 10 expression was observed when the keratinocytes were grown in SG+GF (Fig. 6d), but a significant decrease (p < 0.05) in keratin 10 expression was observed when keratinocytes were grown on irradiated dermal fibroblasts compared with those grown on i3T3. A representative experiment is depicted in Figure 6.

Epidermal formation by cultured keratinocytes

The ability to grow keratinocytes in a completely animal-free culture system may provide a safer technique for producing epidermal grafts for the treatment of wounds. However, for these keratinocytes to be useful for clinical applications they must retain their ability to form a stratified squamous epithelium. To examine whether keratinocytes grown in the SG+GF medium on irradiated human fibroblasts maintained their ability to form a stratified epidermal layer, keratinocytes were seeded onto pieces of decellularized DED. Keratinocyte cultures that had been preestablished on either i3T3 or irradiated HDFs in Green's or SG+GF medium were all shown to form a fully stratified squamous epithelium (Fig. 7). Of note, keratinocytes grown on dermal fibroblasts demonstrated greater levels of stratification compared with those grown on i3T3 in Green's medium (Fig. 7b, e compared with Fig. 7a, respectively). To verify the quality of the reconstructed epidermis, we utilized immunohistochemical analysis to assess the expression of p63, keratin 1/10/11, and keratin 14. Keratinocyte cultures that had been preestablished on either i3T3 or irradiated HDFs in Green's or SG+GF medium were all shown to express p63, keratin 1/10/11, and keratin 14 (Fig. 8). This also revealed that keratinocytes grown on irradiated human fibroblasts showed similar ability to reconstitute the epidermis when compared with keratinocytes grown on i3T3 (Fig. 8a–h). In addition, this analysis demonstrated that keratinocytes that had been grown in serum-free media showed similar ability to reconstitute the epidermis when compared with keratinocytes grown in Green's medium (Fig. 8i–p). The immunoflourescent staining for p63 is indicative of proliferating basal keratinocytes within the epidermis, keratin 1 staining indicates differentiating keratinocytes within the spinosum and granular layers of the epidermis, and keratin 14 staining indicates basal keratinocytes within the reconstructed epidermis.

FIG. 7.

Hematoxylin and eosin staining of reconstituted epidermis produced by seeding cultured keratinocytes onto decellularized human dermis. Keratinocytes previously expanded on i3T3 cells in Green's medium (a) or SG+GF medium (d) were shown to reform a fully stratified epidermal layer. Keratinocytes previously expanded on irradiated human dermal fibroblasts (fib121) both in Green's (b) and SG+GF (e) media were also shown to maintain their ability to reform an epidermal layer. When Green's medium or SG+GF medium was added without keratinocytes (c and f, respectively) no epidermal layer formed (negative control). Scale bar represents 100 μm. Representative images of reconstituted epidermis from three different patients are shown. Color images available online at www.liebertonline.com/ten.

FIG. 8.

Immunohistochemical staining of reconstituted epidermis. Keratinocytes grown on i3T3 cells in Green's medium stained with p63 (a), keratin 1 (b), or keratin 14 (c). Keratinocytes expanded on irradiated human dermal fibroblasts in Green's medium stained with p63 (e), keratin 1 (f), or keratin 14 (g). Keratinocytes grown on either i3T3 cells or irradiated human dermal fibroblasts cells in SG+GF medium stained with p63 (i and m, respectively), keratin 1 (j and n, respectively), or keratin 14 (k and o, respectively). Omission of the primary antibody was used as the negative control (d, h, l, p). Scale bar represents 100 μm. Representative images stained reconstituted epidermis from three different patients are shown. Color images available online at www.liebertonline.com/ten.

Discussion

Although cultured keratinocytes have previously been used for various skin therapies,^1–4 the fact that animal-derived products are still required for their expansion greatly increases the risk associated with these clinical treatments. In view of this, many researchers have investigated ways to reduce these risks by eliminating the need for serum in keratinocyte culture media. In these studies, serum has often been replaced with other supplements such as bovine or porcine pituitary extract.^12,13 However, these supplements offer no major advantages as they are not defined and are of animal origin. Other factors, such as purified bovine serum albumin, have also been used to replace serum¹⁰; however, this is also not fully defined and does not address the risks associated with using animal-derived products.

Because of the potential problems associated with the use of serum alternatives, researchers in our laboratory have recently developed a completely defined medium using combinations of VN, EGF, IGF-binding protein-5, and IGF-I and this medium has been demonstrated to support the isolation and expansion of keratinocytes in the presence of murine i3T3 feeder cells.^14,17 Despite this medium eliminating the need for serum, there are still risks associated with the use of animal-derived feeder cells. Several investigations have shown that human-derived feeder cells are able to support keratinocytes as effectively as murine i3T3 cells.^9,11 For example, Bullock et al. demonstrated that keratinocytes could be expanded by culturing with irradiated HDFs.⁹ This study revealed that these irradiated human fibroblasts were equally effective in supporting initial keratinocyte expansion as the irradiated murine fibroblasts. Further, they demonstrated that the keratinocytes maintained a less differentiated phenotype when cultured in serum-free media with fibroblasts compared with those cultured in serum-supplemented media. However, Bullock et al. isolated both the keratinocytes and fibroblasts using serum, thereby introducing a potential point of contamination.⁹

Because of the potential problems associated with using serum to isolate fibroblast feeder cells, we investigated techniques to isolate and expand primary fibroblasts and keratinocytes in a fully defined serum-free medium. Several growth factors have been shown to effect fibroblast growth and proliferation. It has been demonstrated that the addition of bFGF and/or EGF to serum-free media causes an increase in cell number, which can be further enhanced by the addition of insulin.^29,30 Further, Weinstein et al. have previously shown that serum-free medium containing EGF and insulin could support fibroblast growth.³¹ This medium, however, was supplemented with ovalbumin and a serum-based medium was still required for the isolation and initial expansion of the fibroblasts. Research in our laboratory has also demonstrated that smaller doses of IGF-I can substitute for the larger volumes of insulin required in keratinocyte culture.¹⁴ As such, it is feasible that the addition of IGF-I to fibroblast culture may have the same effect as the addition of insulin. Indeed, it has been demonstrated that the addition of IGF-I to dermal fibroblast culture leads to an increase in cellular proliferation and collagen synthesis.^30,32 Further, the extracellular matrix protein, VN, can induce fibroblast migration using a wound scratch model.³³

With this in mind, we hypothesized that the addition of EGF, bFGF, and our chimeric VN:IGF-I protein²⁰ to basal keratinocyte medium may allow the serum-free expansion of primary fibroblasts. Indeed, as reported herein, a completely defined medium consisting of 50 ng/mL EGF, 50 ng/mL bFGF, 0.5 μg/mL chimeric VN:IGF-I, and 1 mg/mL recombinant HA (SG+FIB medium) was shown to support the isolation and expansion of human primary fibroblasts and HFF cell lines (Figs. 1 and 2 and Supplemental Table S1). Further, these cells could be frozen and stored in liquid nitrogen using the serum-free freezing mix from In Vitro Technologies, thus allowing large stocks to be created for future use. Perhaps the most significant finding, however, was the discovery that primary fibroblasts could be isolated in serum-free conditions by adding at least 5 mg/mL recombinant HA to the collagenase solution.

The ability to isolate and expand fibroblasts in serum-free conditions has eliminated one of the potential problems previously associated with the use of these cells as feeder cell layers for expansion of human primary keratinocytes. We therefore examined whether these fibroblasts were able to support the expansion of keratinocytes using a variation of the SG+FIB medium (SG+GF medium; Figs. 3 and 4 and Supplemental Table S1). This was a logical extension of our study because numerous reports have investigated the effect of EGF on keratinocytes, with several showing that EGF sustains keratinocyte survival^24,34 and reduces keratinocyte cell size.³⁵ bFGF has also previously been demonstrated to support clonal growth of keratinocytes in serum-free medium supplemented with bovine pituitary extract,²⁹ as well as influence keratinocyte migration.²⁶ Moreover, IGF-I enhances keratinocyte migration³⁵ and stimulates keratinocyte proliferation.²¹ It should be noted that future work would investigate the removal of trypsin in the keratinocyte extraction procedure and replace it with a product of nonanimal origin (i.e., TRYPLE). This would eliminate the last remaining potential source of animal-derived contamination.

Analysis of markers expressed by keratinocytes expanded in our serum-free medium indicated that this medium is effective at supporting the isolation and expansion of keratinocytes in an undifferentiated, highly proliferative state (Fig. 5 and 6). Thus, it was demonstrated through both immunofluorescence and real-time PCR analysis that keratinocytes expanded in serum-free conditions maintained expression of p63 and keratin 6. This indicates that our serum-free medium is able to select for highly proliferative, basal keratinocytes. Immunofluorescence analysis also revealed that the keratinocytes maintained minimal immunoreactivity toward the differentiation markers keratins 1 and 10. However, real-time PCR analysis revealed that keratinocytes grown in serum-free conditions had increased expression of keratin 1 but displayed no significant increase in keratin 10 expression. As such, it is likely that our medium is effective at maintaining keratinocytes in an undifferentiated state. We have also demonstrated that irradiated dermal fibroblasts are effective, if not better, than murine i3T3 cells at maintaining keratinocytes in a highly proliferative state. Keratinocytes grown on irradiated dermal fibroblasts were shown to have a lower expression of keratin 1 and 10 mRNA compared with those grown on i3T3, suggesting that these keratinocytes have a less differentiated phenotype. Moreover, we demonstrated that keratinocytes grown in serum-free conditions on irradiated dermal fibroblasts maintained their ability to form a fully stratified squamous epithelial layer using an in vitro HSE model (Fig. 7). We confirmed that the reconstituted epidermis contained keratinocytes that expressed p63 and keratin 14 in the proliferating basal layer and keratin 1/10/11 in the stratified layer (Fig. 8). Taken together, these findings indicate that the SG+GF medium with the HDF feeder cells is suitable for growing keratinocytes for clinical use.

In conclusion, the development of a serum-free culture system for primary keratinocytes represents a significant improvement to the current “gold standard” technique of Green's serum-supplemented medium and murine feeder cells and will likely allow safer expansion of these cells for clinical use. The full validation of our serum-free culture method for clinical applications would be the next step following the proof-of-concept research documented in this article. In addition to the therapeutic advantages gained through growing keratinocytes in a defined medium, this medium will also facilitate fundamental investigative studies into the paracrine interactions between keratinocytes and fibroblasts. For example, the use of a fully defined, minimal protein content, serum-free medium may provide a significant advantage for these studies as the absence of serum in the medium means that the critical factors secreted by the fibroblasts, which may be important for supporting keratinocyte cell growth, will be more readily detected.

Footnotes

Acknowledgments

The authors thank Ms. Rebecca Dawson for collecting the skin samples and for her support and training in tissue culture techniques, Mr. D. Geyer for assistance with the histology, and Daniel Broszczak for performing the immunohistochemical staining of the reconstituted skin. Thanks are also extended to the Royal Children's Hospital, St. Andrews and Wesley Hospitals for donation of skin samples and Red Cross Blood Services for their irradiating services. The authors acknowledge Tissue Therapies Ltd. and the Golden Casket Lottery Corporation for their financial support.

Disclosure Statement

No competing financial interests exist.

References

Wood

F.M.

, Stoner

Implication of basement membrane development on the underlying scar in partial-thickness burn injury. Burns (J Int Soc Burn Injuries), 22:459. 1996.

De Luca

, Albanese

, Bondanza

, Megna

, Ugozzoli

, Molina

, Cancedda

, Santi

P.L.

, Bormioli

, Stella

, Magliacani

Multicentre experience in the treatment of burns with autologous and allogenic cultured epithelium, fresh or preserved in a frozen state. Burns (J Int Soc Burn Injuries), 15:303. 1989.

Moustafa

, Simpson

, Glover

, Dawson

R.A.

, Tesfaye

, Creagh

F.M.

, Haddow

, Short

, Heller

, MacNeil

A new autologous keratinocyte dressing treatment for non-healing diabetic neuropathic foot ulcers. Diabet Med, 21:786. 2004.

de Luca

, Albanese

, Megna

, Cancedda

, Mangiante

P.E.

, Cadoni

, Franzi

A.T.

Evidence that human oral epithelium reconstituted in vitro and transplanted onto patients with defects in the oral mucosa retains properties of the original donor site. Transplantation, 50:454. 1990.

Rheinwald

J.G.

, Green

Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell, 6:331. 1975.

Rolleston

W.B.

Bovine serum: reducing the variables through the use of donor herds. Dev Biol Stand, 99:79. 1999.

Heiskanen

, Satomaa

, Tiitinen

, Laitinen

, Mannelin

, Impola

, Mikkola

, Olsson

, Miller-Podraza

, Blomqvist

, Olonen

, Salo

, Lehenkari

, Tuuri

, Otonkoski

, Natunen

, Saarinen

, Laine

N-glycolylneuraminic acid xenoantigen contamination of human embryonic and mesenchymal stem cells is substantially reversible. Stem Cells, 25:197. 2007.

Martin

M.J.

, Muotri

, Gage

, Varki

Human embryonic stem cells express an immunogenic nonhuman sialic acid. Nat Med, 11:228. 2005.

Bullock

A.J.

, Higham

M.C.

, MacNeil

Use of human fibroblasts in the development of a xenobiotic-free culture and delivery system for human keratinocytes. Tissue Eng, 12:245. 2006.

10.

Castro-Munozledo

, Hernandez-Quintero

, Marsch-Moreno

, Kuri-Harcuch

Cultivation, serial transfer, and differentiation of epidermal keratinocytes in serum-free medium. Biochem Biophys Res Commun, 236:167. 1997.

11.

Sun

, Higham

, Layton

, Haycock

, Short

, MacNeil

Developments in xenobiotic-free culture of human keratinocytes for clinical use. Wound Repair Regen, 12:626. 2004.

12.

Wang

H.J.

, Chen

T.M.

, Cheng

L.F.

, Cheng

T.Y.

, Tung

Y.M.

Human keratinocyte culture using porcine pituitary extract in serum-free medium. Burns (J Int Soc Burn Injuries), 21:503. 1995.

13.

Boyce

S.T.

, Ham

R.G.

Cultivation, frozen storage, and clonal growth of normal human epidermal keratinocytes in serum-free media. Methods Cell Sci, 9:83. 1985.

14.

Dawson

R.A.

, Upton

, Malda

, Harkin

D.G.

Preparation of cultured skin for transplantation using insulin-like growth factor I in conjunction with insulin-like growth factor binding protein 5, epidermal growth factor, and vitronectin. Transplantation, 81:1668. 2006.

15.

Coolen

N.A.

, Verkerk

, Reijnen

, Vlig

, van den Bogaerdt

A.J.

, Breetveld

, Gibbs

, Middelkoop

, Ulrich

M.M.

Culture of keratinocytes for transplantation without the need of feeder layer cells. Cell Transplant, 16:649. 2007.

16.

Horch

R.E.

, Debus

, Wagner

, Stark

G.B.

Cultured human keratinocytes on type I collagen membranes to reconstitute the epidermis. Tissue Eng, 6:53. 2000.

17.

Richards

, Leavesley

, Topping

, Upton

Development of defined media for the serum-free expansion of primary keratinocytes and human embryonic stem cells. Tissue Eng Part C Methods, 14:221. 2008.

18.

Higham

M.C.

, Dawson

, Szabo

, Short

, Haddow

D.B.

, MacNeil

Development of a stable chemically defined surface for the culture of human keratinocytes under serum-free conditions for clinical use. Tissue Eng, 9:919. 2003.

19.

Meana

, Iglesias

, Del Rio

, Larcher

, Madrigal

, Fresno

M.F.

, Martin

, San Roman

, Tevar

Large surface of cultured human epithelium obtained on a dermal matrix based on live fibroblast-containing fibrin gels. Burns, 24:621. 1998.

20.

Van Lonkhuyzen

D.R.

, Hollier

B.G.

, Shooter

G.K.

, Leavesley

D.I.

, Upton

Chimeric vitronectin:insulin-like growth factor proteins enhance cell growth and migration through co-activation of receptors. Growth Factors, 25:295. 2007.

21.

Barreca

, De Luca

, Del Monte

, Bondanza

, Damonte

, Cariola

, Di Marco

, Giordano

, Cancedda

, Minuto

In vitro paracrine regulation of human keratinocyte growth by fibroblast-derived insulin-like growth factors. J Cell Physiol, 151:262. 1992.

22.

Harrison

C.A.

, Gossiel

, Bullock

A.J.

, Sun

, Blumsohn

, Mac Neil

Investigation of keratinocyte regulation of collagen I synthesis by dermal fibroblasts in a simple in vitro model. Br J Dermatol, 154:401. 2006.

23.

Ainscough

S.L.

, Barnard

, Upton

, Harkin

D.G.

Vitronectin supports migratory responses of corneal epithelial cells to substrate bound IGF-I and HGF, and facilitates serum-free cultivation. Exp Eye Res, 83:1505. 2006.

24.

Rheinwald

J.G.

, Green

Epidermal growth factor and the multiplication of cultured human epidermal keratinocytes. Nature, 265:421. 1977.

25.

DeLapp

N.W.

, Dieckman

D.K.

Effect of basic fibroblast growth factor (bFGF) and insulin-like growth factors type I (IGF-I) and type II (IGF-II) on adult human keratinocyte growth and fibronectin secretion. J Investig Dermatol, 94:777. 1990.

26.

Sogabe

, Abe

, Yokoyama

, Ishikawa

Basic fibroblast growth factor stimulates human keratinocyte motility by Rac activation. Wound Repair Regen, 14:457. 2006.

27.

Tsuboi

, Sato

, Shi

C.M.

, Ogawa

Stimulation of keratinocyte migration by growth factors. J Dermatol, 19:652. 1992.

28.

Ristow

H.J.

, Messmer

T.O.

Basic fibroblast growth factor and insulin-like growth factor I are strong mitogens for cultured mouse keratinocytes. J Cell Physiol, 137:277. 1988.

29.

Shipley

G.D.

, Keeble

W.W.

, Hendrickson

J.E.

, Coffey Robert

, Pittelkow

M.R.

Growth of normal human keratinocytes and fibroblasts in serum-free medium is stimulated by acidic and basic fibroblast growth factor. J Cell Physiol, 138:511. 1989.

30.

Narine

, De Wever

, Van Valckenborgh

, Francois

, Bracke

, DeSmet

, Mareel

, Van Nooten

Growth factor modulation of fibroblast proliferation, differentiation, and invasion: implications for tissue valve engineering. Tissue Eng, 12:2707. 2006.

31.

Weinstein

, Hoover

G.A.

, Majure

, van der Spek

, Stemerman

M.B.

, Maciag

Growth of human foreskin fibroblasts in a serum-free, defined medium without platelet-derived growth factor. J Cell Physiol, 110:23. 1982.

32.

Rolfe

K.J.

, Cambrey

A.D.

, Richardson

, Irvine

L.M.

, Grobbelaar

A.O.

, Linge

Dermal fibroblasts derived from fetal and postnatal humans exhibit distinct responses to insulin like growth factors. BMC Dev Biol, 7:124. 2007.

33.

Lygoe

K.A.

, Wall

, Stephens

, Lewis

M.P.

Role of vitronectin and fibronectin receptors in oral mucosal and dermal myofibroblast differentiation. Biol Cell, 99:601. 2007.

34.

Jost

, Huggett

T.M.

, Kari

, Rodeck

Matrix-independent survival of human keratinocytes through an EGF receptor/MAPK-kinase-dependent pathway. Mol Biol Cell, 12:1519. 2001.

35.

Haase

, Evans

, Pofahl

, Watt

F.M.

Regulation of keratinocyte shape, migration and wound epithelialization by IGF-1- and EGF-dependent signalling pathways. J Cell Sci, 116:3227. 2003.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.00 MB