TISSUE ENGINEERING
Volume 13, Number 11, 2007
# Mary Ann Liebert, Inc.
DOI: 10.1089/ten.2006.0329
Generation of Human Epidermal Constructs
on a Collagen Layer Alone
FEDERICA RIVA, Ph.D.,1 ANDREA CASASCO, M.D.,1 EMANUELE NESPOLI, M.D.,2
ANTONIA ICARO CORNAGLIA, Ph.D.,1 MARCO CASASCO, M.D.,1 ANGELA FAGA, M.D.,2
SILVIA SCEVOLA, M.D.,2 GIULIANO MAZZINI, Ph.D.,3 and ALBERTO CALLIGARO, Ph.D.1
ABSTRACT
Because engineered tissues are designed for clinical applications in humans, a major problem is the contamination of cocultures and tissues by allogenic molecules used to grow stem cells in vitro. The protocols
that are commonly applied to generate epidermal equivalents in vitro require the use of irradiated murine
fibroblasts as a feeder layer for keratinocytes. In this study, we report a simple procedure for growing
human keratinocytes, isolated from adult skin, to generate an epidermal construct on a collagen layer alone.
In this model, no human or murine feeder layers were used to amplify cell growth, and isolated keratinocytes were seeded directly at high cell density on the collagen-coated flasks or coverslips in an epithelial
growth medium containing low calcium concentration.
Morphological, immunochemical, and cytokinetic features of epithelial colonies grown on the collagen
layer were typical of keratinocytes and were comparable with those reported for keratinocytes grown on
a feeder layer. The stratification of keratinocytes generated 3-dimensional synthetic constructs displaying
a tissue architecture comparable with that of natural epidermis. Epithelial cells expressed specific markers
of keratinocyte terminal differentiation, including involucrin and filaggrin. Nevertheless, the number of cell
layers was lower than in natural skin, and electron microscopical analysis revealed that the overall organization of these layers was poor compared with natural epidermis, including the formation of junctional
complexes, basement membrane, and keratinization. The lack of epithelial–mesenchymal interactions that
occur during skin histogenesis may account for such an incomplete maturation of epidermal constructs.
INTRODUCTION
A
and culturing technology have provided the basis for in vitro organ reconstruction, in particular skin substitutes. First, long-term
subcultivation of keratinocytes permitted the growth and
differentiation of normal human keratinocytes in culture.1
Subsequently, the coculture of epidermal keratinocytes and
dermal fibroblasts permitted the generation of different
models of tissue-engineered skin (TES).2–15 An alternative
DVANCES IN STEM CELL CLONING
protocol is the serum-free method,3,16,17 which has permitted the production of 3-dimensional (3D) epidermal equivalents on synthetic polycarbonate filters.18,19
TES can be regarded as the only engineered organ that is
currently used in clinical applications, including the treatment of burns and chronic wounds, in which the need to
accelerate healing seems to be a major goal.20
Models of bilayered TES consist of dermal and epidermal equivalents, thus reconstructing the entire skin architecture. Epidermal and dermal cells used to produce skin
1
Department of Experimental Medicine, Histology and Embryology Unit, and 2Department of Plastic Surgery, University of Pavia,
Pavia Italy; and 3Department of Animal Biology, Institute of Molecular Genetics IGM-CNR, Histochemistry and Cytometry Unit,
Pavia, Italy.
1 (page numbers are temporary)
RIVA ET AL.
2
constructs can be isolated from different sources, including
autogeneic, allogeneic, and xenogeneic skin.21,22 Moreover,
it has been possible to introduce melanocytes, Langerhans
cells, blood vessels, and hair in some models of engineered
skin.23–27 Different 3D scaffolds and biomatrices have been
used to promote keratinocyte and fibroblast growth.20 These
include collagen,2,7,8,10,28 glycosaminoglycans such as benzyl
ester of hyaluronan,29 and synthetic biopolymers.18,19
According to Rheinwald and Green,1 the protocol that is
widely used to seed and clone epidermal cells requires the
use of a feeder layer of mitotically inhibited murine fibroblasts, namely, lethally irradiated 3T3 murine fibroblasts,
which acts as an adhesion surface.30 However, the use of
a xenogenic cell type represents a major drawback of this
technique, and it can be argued that the use of xenogenic
cells requires the addition of antigenic components and presents the risk that viruses, prions, or other macromolecules
may be transmitted to human cells from mouse cells during
tissue culture. This should be taken into account in the case
that TES is designed for clinical application.12,29
To find an alternative method for keratinocyte cultivation
without the use of a murine feeder layer, we have studied
the possible application of a film of bovine type I collagen,
an acellular natural polypeptide biomaterial, as a support for
cell growth in a defined culture medium containing bovine
serum and low calcium concentration.31,32 In this study, we
demonstrated that human keratinocytes, isolated from adult
skin, are able to grow and proliferate for several generations
on a collagen layer alone and to achieve terminal differentiation without using a feeder layer or increasing calcium
concentration in the culture medium.33
MATERIALS AND METHODS
Cell culture
Skin biopsies (n ¼ 30) were obtained from clinically
healthy human skin from different body areas (abdomen,
thigh, breast) from donors of different ages (range 40–60
years). Human epidermal keratinocytes were isolated according to Häkkinen et al.34 Surgical biopsies (*44 cm)
were cut into smaller fragments (*11 cm) and subsequently digested with 2.5 mg/mL Protease type IX (SigmaAldrich, St. Louis, MO) at 48C for 24 h to separate the
epidermis from the dermal layer. The epidermal sheet was
carefully peeled from the underlying dermis and trypsinized
(0.05% trypsin, 0.01% ethylenediaminetetraacetic acid) for
10 min at room temperature, to dissociate single cells.
Isolated cells were centrifuged (200 g100 at 158C), and
the pellet of keratinocytes (*5107 total cells) was gently
re-suspended in the growth medium. Cells were seeded in
culture flasks (25 cm2) or on coverslips (12 mm ø) previously coated with bovine collagen type I (Sigma, 10 mg of
calf skin in 2 mL of 0.1-M acetic acid and 8 mL of sterile
water) at a final seeding cell density of approximately 10 to
20106 viable per well and 2105 per coverslip. The
cultures were incubated at 378C in 5% carbon dioxide with
cell epidermal culture medium (CEC) without epidermal
growth factor (EGF) and with low Ca2þ concentration.35
CEC medium is a 3:1 mixture of Dulbecco’s modified
Eagle medium (Sigma) and Ham’s F12 (Sigma) containing
10% fetal bovine serum (Eurobio, Courtaboeuf, France),
0.4 mg/mL hydrocortisone (Sigma), 5 mg/mL insulin (Sigma),
5 mg/mL Apo-transferrin (Sigma), 1.36 ng/mL triiodo-Lthyronine (Sigma), 1 mg/mL 200-mM L-glutamine (Eurobio),
0.11 mg/mL 100-mM sodium pyruvate (Sigma), 100 mg/mL
penicillin (Eurobio), and 100 mg/mL streptomycin (Eurobio).
Both media contain a low (0.03-mM) concentration of Ca2þ.
After 24 h of culture, 10 ng/mL EGF (human recombinant,
Sigma) was added to the cultures. Complete fresh CEC medium (CEC medium with EGF (CECþ)) has a pH of 8.1, but
it diminished to 6.5 after 2 to 3 days of culture; therefore, it
was changed every 2 days. After different time steps, cells
were fixed and processed for immunocytochemistry. To
evaluate the effect of bovine serum on keratinocyte growth
and differentiation in our experimental model, we also seeded
the cells in a serum-free medium.
Immunocytochemistry
Cells seeded on coverslips were fixed with 4% paraformaldehyde in 0.1-M phosphate buffer (pH 7.4) for 15 min
and postfixed in 70% ethanol for 30 min. Immunostaining
was performed by incubating the coverslips for 1 h with
primary monoclonal antibodies to epithelial and nonepithelial markers, followed by a secondary antimouse antibody
(fluorescein isothiocyanate (FITC)-conjugated antibody,
Sigma; Alexa 594, Molecular Probes–Invitrogen Ltd, Glasgow, Scotland, UK). Nuclear staining of deoxyribonucleic
acid (DNA) with Hoechst 33258 dye (0.5 mg/mL) was also
performed. The fluorescence images were obtained using
a Zeiss Axiophot fluorescence microscope (Carl Zeiss,
Oberkochen, Germany) and a Leica TCS SP2 confocal
microscope (Heidelberg, Germany), by acquiring green and
red fluorescence signals at 0.3-mm intervals. Image analysis
was performed using the Leica Confocal Software.
Immunohistochemistry
Three-dimensional epidermal constructs obtained from
primary cultures were gently detached from the flask using a
scalpel, fixed with 4% paraformaldehyde in 0.1-M phosphate
buffer (pH 7.4) for 6 h, dehydrated through graded concentrations of ethanol, and embedded in paraffin. Sections were
obtained at 5 to 10 mm, rehydrated, and stained with Harris’s
haematoxylin and eosin or processed for immunohistochemical staining according to the indirect streptavidin-biotin
immunoperoxidase technique. The list of cell differentiation–
and cell proliferation–related antigens that have been investigated is shown in Table 1.
Briefly, the sections were incubated serially with the
following solutions: 0.3% hydrogen peroxide for 30 min to
EPIDERMAL CONSTRUCT ON A COLLAGEN LAYER
TABLE 1.
Primary antibody to
pan cytokeratin
(clone LU5, Bachem, UK)
vimentin
(clone V9, Biogenex, USA)
filaggrin
(clone FLG01, Oncogene, USA)
p63
(clone 4A4, Oncogene, USA)
involucrin
(SY5, Sigma, USA)
laminin
(clone LAM-89, Sigma, USA)
integrin b1 subunit
(clone JB1 a ( J-10), Biogenex, USA)
a2b1 integrin
(clone BHA2.1, Chemicon, USA)
5-bromo-20 deoxyuridine
(clone Bu-1, Amersham Biosc., UK)
3
DETAILS OF PRIMARY ANTIBODIES
Immunoperoxidase
(dilution)
Pre-digestion
with
protease
of paraffin
sections
Pre-treatment
with
Microwave
of paraffin
sections
Immunofluorescence
(dilution)
1:400
þ
1:200
1:400
1:100
1:200
þ
1:100
1:1000
þ
1:100
1:200
1:100
1:2000
þ
1:200
1:100
þ
1:100
1:50
þ
1:200
1:500*
1:100*
*Enzyme pre-digestion with nuclease or pre-treatment with HCl
remove endogenous peroxidase activity normal goat serum,
diluted 1:20, for 30 min to reduce background staining primary antisera or antibodies at appropriate dilutions (range
1:100–1:2000), overnight at 48C biotinylated goat antirabbit or anti-mouse immunoglobulin G (Super Sensitive
kit, BioGenex, San Ramon, CA) for 1 h at room temperature
streptavidin-biotinylated peroxidase complexes (Super Sensitive kit, BioGenex) for 1 h at room temperature 0.03%
3,30 -diaminobenzidine tetrahydrochloride, to which hydrogen peroxide (0,02%) was added just before use, for 5 min
at room temperature.
Some sections were lightly counterstained with hematoxylin. Finally, the sections were dehydrated, mounted with DPX
mounting medium, and examined. Each solution was prepared in 0.05-M Tris buffer (pH 7.4) containing 0.1-M sodium
chloride (0.15-M Tris-buffered saline), and between each of
the steps of the immunostaining procedure, the sections were
washed in the same buffer. Before immunostaining, some
sections were pretreated with protease (0.5% pepsin in 0.01M hydrochloric acid (HCl) or 0.1% trypsin in distilled water,
pH 7.6) for 10 to 30 min at 378C. Other sections were incubated in a microwave oven for antigen retrieval, according to
the antibody supplier’s protocol. Microwave heating was performed in an antigen retrieval solution (BioGenex) as irradiation fluid by using a Panasonic microwave oven (model
4540, 900 W), as described previously.36 Specificity controls
included omission of the primary antibody the substitution
of nonimmune sera for the primary antibody. Immunohistochemistry on biopsies of natural skin and epidermis was
performed to compare the patterns of immunostaining.
Transmission electron microscopy
Samples of epidermal constructs were immersed in a
solution containing 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1-M sodium cacodylate buffer (pH 7.4) for
6 h. They were then postfixed for 1 h in 1.33% osmium tetroxide in 0.1-M collidine buffer. Tissues were dehydrated
through graded alcohols and embedded in Epon 812. Semithin sections were stained with toluidine blue. Thin sections
were stained with uranyl acetate and lead citrate and observed using a Zeiss EM10 electron microscope.
Flow cytometry
Freshly isolated cells obtained after trypsinization treatment, were fixed with 70% ethanol at 208C for 10 min
and conserved at 208C. The other cells grown on collagen
layer were maintained in culture for different lengths of
time (1, 7, 10, 15, and 30 days) after isolation, and during
the last 2 h of culture, 30-mM of 5-bromo-20 -deoxyuridine
(BrdU, B-5002, Sigma) was added to the medium. Finally,
the samples were washed in phosphate buffered saline (PBS)
and fixed in 70% ethanol. Incorporated BrdU was detected
using immunostaining with monoclonal anti-BrdU antibody
(Amersham Biosciences, Buckinghamshire, UK). Briefly,
fixed cells were removed from the fixative, washed in PBS,
centrifuged, and pellet treated with 2 N HCl for 30 min at
room temperature. Cells spun down were neutralized with
0.1-M sodium tetraborate (pH 8.5) for 15 min, centrifuged,
and washed in PBS containing 1% bovine serum albumin
and 0.2% Tween 20 (PBT) solution. After centrifugation,
RIVA ET AL.
4
the cells were incubated in the same PBT solution for 15 min
to block unspecific binding of antibodies. Cells (*2106
per sample) were then incubated with monoclonal anti-BrdU
antibody (diluted 1:100 in PBT) for 1 h at room temperature. At the end of this period, samples were washed 3 times
(10 min each) in PBS and incubated for 30 min in PBT
containing FITC-conjugated antimouse antibody (diluted
1:100). Thereafter, cells were washed twice in PBT and
finally re-suspended in PBS. For biparametric analysis of
immunofluorescence versus DNA content, at the end of the
immunological reaction, cells were resuspended in PBS
containing propidium iodide (5 mg/mL) and RNase Type 1A
(1 mg/mL, Sigma). Samples were analyzed after at least
30 min of staining at room temperature or kept overnight at
48C and measured the next day. Cells were analyzed with a
Partec PAS II flow cytometer (Partec, Munster, Germany).
Assessment of cell distribution in each phase of the cell
cycle was evaluated according to flow cytometric analysis
of nuclear DNA content performed by the software provided with the instrument.
FIG. 1. Immunostaining for vimentin (A, B) and cytokeratins
(C, D) on fragments of epidermis and dermal connective tissue
isolated from natural skin after enzymatic treatment of protease
IX. Vimentin and keratins are specifically expressed in dermal
fibroblast and epidermal keratinocytes, respectively. Scale bar:
20 mm.
RESULTS
Isolation of epidermal cells
To ensure that only epidermal keratinocytes were isolated and cloned, using immunohistochemical techniques, we
analyzed bio-optical fragments of dermal connective tissue
and epidermis isolated after enzymatic treatment of Protease IX. Immunostaining for vimentin and cytokeratins on
these fragments showed that specific proteins were expressed in dermal fibroblasts and epidermal keratinocytes, respectively (Fig. 1). Epidermal cells were vimentin negative
and showed keratin expression (Fig. 1A, C), and dermal
cells displayed vimentin immunoreactivity and were cytokeratin negative (Fig. 1B, D).
Keratinocytes were isolated from an epidermal fragment
that has been shown to be composed exclusively of cytokeratin-positive cells and displayed no vimentin staining,
thus showing that dermal connective cells were absent from
the fragment. Subsequently, keratinocytes were seeded on a
type I collagen layer in the culture flask.
epithelial cells established interactions using cytoplasmic
protrusions and finally several colonies, thus forming a
continuous epithelial monolayer. At 30 days, these cells
constituted a stratified sheet made of 4 to 6 cell layers on
the collagen matrix (3D epithelial constructs, Fig. 2C, D).
The formation of keratinocytes colonies and a stratified
sheet was not obtained in serum-free cultures.
Cell culture and formation of epidermal constructs
As soon as 7 days, primary epidermal keratinocytes isolated from adult skin formed visible colonies, and each
colony was the progeny of a single cell or small cell cluster
of 10 to 20 epidermal cells (Fig. 2A). The colonies in the
primary cultures formed mosaics of polygonal and squamous cells, thus suggesting an epithelial origin (Fig. 2B).
Accordingly, all cells showed cytoplasmic immunostaining for cytokeratins, and 10% of cultured keratinocytes
expressed the nuclear protein p63 (Fig. 3). After 10 days,
cells seeded on the collagen layer grew, forming large
clones with high proliferative activity. Within the colonies,
FIG. 2. Primary culture of human keratinocytes seeded on flasks
coated with type I collagen. Seven days after isolation, a single
cell or a small cell cluster start to form clones (A), which are
microscopically visible as colonies of 10 to 20 cells 10 days later
(B). After 4 weeks, these clones constituted in vitro an artificial
epidermal tissue of 4 to 6 cell layers (C) (the 3-dimensional epithelial construct visible inside the culture flask (D)). Scale bars: (B, C)
20 mm.
EPIDERMAL CONSTRUCT ON A COLLAGEN LAYER
5
tion, the number of cycling cells (S, G2, and M cells) increased to 20% to 30% of seeded cells (Fig. 4B, D).
No significant differences in cultured keratinocytes were
found in cells isolated from donors of different ages or from
skin from different body areas.
Histological and immunohistochemical analyzes
FIG. 3. Human keratinocytes seeded on collagen-coated coverslips. All cells show cytoplasmic immunostaining for pancytokeratins (A). Some keratinocytes express the protein p63, a
nuclear marker associated with putative stem cells of ectodermderived epithelial tissues (B). Scale bars: (A, B) 20 mm.
Epidermal constructs were made using a stratified epithelium made with 4 to 6 cell layers and thus thinner than
natural epidermis (Fig. 5). It was possible to identify cell
layers comparable with those observed in natural skin
(basal, suprabasal, and superficial layers). In the basal layer,
epithelial cells were cuboidal cells, whereas in the other
layers, cells were flat. In the suprabasal layer, cells displayed
cytoplasmic protrusions similar to those observed in the
spinous layer of natural skin.
Immunostaining of the constructs revealed that epithelial
cells were positive for several proliferation and differentiation markers (Fig. 5). All cells were cytokeratin positive
Proliferation
Immunofluorescence and flow cytometric analyses revealed that, after 1 h of incubation, the percentage of isolated cells that had incorporated bromodeoxyuridine was
less than 1%, thus showing that cells after seeding were
mostly G0/G1 cells (Fig. 4A, C). After 3 days of incuba-
4500
250
Gate: R1
200
G1
FSC
3600
150
100
50
counts
2700
0
0
1000
50
100 150
FL3
200
250
50
100 150
FSC
200
250
50
100 150 200 250
FL3
Gate: R1
1800
FSC
100
900
10
1
0
S
0
1600
50
G2+M
100
FL3
150
200
0.1
250
0
1000
Gate: R1
100
FSC
1280
G1
10
1
counts
960
0.1
0
1000
640
100
FSC
S
320
G2+M
0
0
50
100
1
150
FL3
10
200
250
0.1
0.1
1
10
SSC
100
1000
FIG. 4. Analysis of bromodeoxyuridine (BrdU) incorporation according to immunofluorescence and flow cytometry. Nuclear BrdU
immunofluorescence in keratinocytes seeded on collagen-coated coverslips at time 0 and after 3 days of culture (A, B). Cell-cycle analysis
of epithelial cells, collected at time 0 and 3 days of culture after isolation (C, D). Cell distribution in the G0/G1, S, G2, and M phases was
determined using flow cytometry on samples counter-stained with propidium iodide. No cells had incorporated BrdU immediately after
seeding (A), thus suggesting that cells were mostly in the G0/G1 phase (C). After 3 days of culture, the number of BrdU-incorporating cells
increased (B), and the number of cycling cells (S, G2, and M cells) increased to 20% to 30% of seeded cells (D).
RIVA ET AL.
6
(Fig. 5D), and all basal cells displayed integrin b1 subunit
membrane staining (Fig. 5F) and a2b1 integrin intracytoplasmic staining (Fig. 5F*), whereas laminin immunoreactivity was cytoplasmic (Fig. 5H). Basal and suprabasal cells
were p63 positive (Fig. 5L). Cells of most superficial layers
displayed involucrin and filaggrin immunostaining (Fig.
5N, P). The constructs were vimentin negative (Fig. 5R).
Laser scanning confocal microscopy analysis
Examination of epidermal equivalents revealed that keratinocytes are organized in multiple layers, comparable with
those observed in natural skin. It was possible to observe an
increasingly cytoplasmic expression of filaggrin, a marker
of keratinocyte differentiation, from the intermediate to the
superficial layer, thus suggesting a progression in cell differentiation (Fig. 6).
Electron microscopy analysis
Electron microscopy analysis confirmed that keratinocytes grew and stratified on the collagen layer, thus forming
a thin epidermal equivalent. Degenerated keratinocytes
were observed detaching from the surface of the constructs
(Fig. 7A). The number of the cell layers observed in synthetic constructs (4–6 layers, including basal cells) was
lower than in natural skin and other models of bilayered
skin equivalents. Using light microscopy analysis, it was
possible to identify cell layers comparable with those in
natural skin. Nevertheless, electron microscopy analysis
revealed that the overall organization of these layers was
poorer than in natural epidermis. In particular, few and
poorly developed desmosomes were observed in the spinous
layer, and few keratohyalin granules occurred in the cytoplasm of the cells of the granular layer (Fig. 7B). All epithelial cells contained abundant tonofibrils throughout their
cytoplasm (Fig. 7C). No basement membrane could be observed underlying the basal layer of epidermal constructs.
DISCUSSION
Different models of TES have been regarded to as useful
tools for studies in skin biology, pharmacology, and toxicology and for the treatment of skin wounds. Our aim was
FIG. 5. Comparison in hematoxylin and eosin and immunohistochemical stainings between natural human epidermis (A, C, E,
E*, G, I, M, O, Q) and synthetic epidermal construct (B, D, F,
F*, H, L, N, P, R). Haematoxylin ad eosin staining (A, B). Immunostaining for pan-cytokeratins (C, D); integrin b1 subunit
(E, F); integrin a2b1 (E*, F*), laminin (G, H); p63 (I, L); involucrin (M, N); filaggrin (O, P). Vimentin immunostaining was
restricted to dermal cells of natural epidermis (Q, R). Scale bars:
(A–R) 50 mm. E, epidermis; EC, epidermal construct.
EPIDERMAL CONSTRUCT ON A COLLAGEN LAYER
FIG. 6. Keratinocyte constructs grown for 10 days on type I
collagen-coated coverslips as observed using laser scanning confocal microscopy. Cells were stained with a primary antibody against
filaggrin, a cytoplasmic protein specific for epidermal differentiation (red). Nuclear deoxyribonucleic acid blue fluorescence using
Hoechst 33258 (blue). Magnification 100. (A) Z-scan images of
the artificial epidermal sheet (sections thickness: 10 mm; 0.3-mm
intervals). Every section represents a different cell layer and shows
filaggrin localization (red fluorescence). (B) x-y cross-section
shows the high expression of filaggrin in the cells of the upper
layer alone. (C) x-y cross-section of the lower layer of the epidermal sheet. The cytoplasmic immunostaining of filaggrin is no
longer evident. (D) x-y cross-section of the basal layer of the
epidermal construct; all cells are small and negative for filaggrin
staining. An orthogonal projection of two cross-sectional cuts (x-z
and y-z) is also shown for the B, C, and D images. Color images
available online at www.liebertpub.com /ten.
to assess minimal experimental conditions that would permit the generation of epidermal constructs in vitro.
Because engineered tissues are designed for clinical applications in humans, a major problem is contamination of
tissues by allogenic molecules or xenogenic cells used to
clone and grow stem cells in vitro. In particular, the pro-
7
FIG. 7. Electron microscopy analysis of an engineered epidermal construct. The ultrastructural organization of cell layers in
the construct is comparable with natural epidermis. (A) In the superficial layers (SL), the cells are flat, and it is possible to observe
cells detaching from the surface of the construct (scale bar: 2 mm).
(B) Poorly developed desmosomes are visible in the spinous layer
(arrows, scale bar: 0.6 mm), and keratohyalin granules are detectable in the granular layer (inset, scale bar: 1.5 mm). (C) Cytoplasmic and nuclear details of an epidermal cell in the basal layer.
Note abundant tonofibrils and tonofilaments (TF) in the cytoplasm.
N, nucleus (scale bar: 1 mm).
tocols that are commonly applied to generate epidermal
equivalents in vitro require the use of irradiated murine
fibroblasts as a feeder layer for keratinocyte cells.1,37 We
used a simple procedure for growing normal human keratinocytes without using any feeder layer. Recently, a feeder
layer made of irradiated human fibroblast was successfully
used to aid the expansion of keratinocytes in vitro,38,39 thus
avoiding the use of xenogenic cells. Although the possible
presence of pathogens in the human feeder layer cannot be
excluded, this strategy may provide the basis for the generation of epidermal equivalents supported by patients’ own
fibroblasts.
The isolated keratinocytes, seeded directly on collagencoated flasks in medium containing a low calcium concentration, grew and proliferated for several generations. The
differentiation of cultured keratinocytes started even without increasing calcium concentration in the culture medium
8
and without adjusting the cultures at the air–liquid interface.40 No significant growth, differentiation, or stratification of keratinocytes could be observed in the cultures
without adding serum to the medium.
The morphological and immunochemical features of epithelial colonies grown on the collagen layer were typical of
keratinocyte cells, forming mosaics of polygonal and squamous cells. Moreover, proliferation dynamics were comparable with those reported for keratinocytes grown on the
murine feeder layer.1,41,42 Previous studies have shown that
epithelial–mesenchymal interactions (epidermal-dermal signaling) are required for epidermal histogenesis in vitro.3,4,43–45
Type I collagen is the major matrix components of dermis
(the mesenchymal counterpart of the epidermis in natural
skin). We assumed that it could be the most representative
matrix that could mimic the connective tissue underlying the
differentiation of keratinocytes in vitro.
Our experiments suggest that type I collagen alone,
which is the major matrix component of the dermis, may be
sufficient to trigger and sustain keratinocyte differentiation
and epidermal histogenesis. Accordingly, no adhesion, differentiation, or stratification was observed in control uncoated polystyrene flasks.
The extensive capacity of the epidermal cell renewal
in vivo depends on a minor population of stem cells located
in hair follicles and the basal layer of adult epidermis.23,46
p63 protein, a recently identified member of the p53 gene
family, seems to be principally restricted to cells with high
proliferative potential and is not expressed in epithelial
cells undergoing terminal differentiation.47 The percentage
of p63-positive cells observed in our cultures after seeding
were similar to the percentage of putative stem cells in the
basal epidermal layer.48,49 It can be presumed that such
cells are able to produce clones having particular stemness
features; bromodeoxyuridine incorporation, as detected using
immunofluorescence and flow cytometry, suggested that
these cells, which were traversing the G1 cell-cycle phase,
were able to enter in S-phase and generate cell clones that
subsequently underwent the differentiation program.
As to the source of keratinocytes, no significant differences in cloning potential could be observed in cells isolated
from donors of different ages or from skin from different
anatomical sites. Our data have demonstrated that keratinocytes stratified on the collagen layer, thus forming epidermal
3D equivalents. Immunohistochemical analysis of the constructs confirmed that all cell layers consisted of cytokeratinpositive epithelial cells,50 whereas no connective tissue cells
could be identified using vimentin-immunostaining. Within
synthetic constructs, it was possible to identify cell layers
comparable with those observed in natural skin; nevertheless,
the number of the cell layers (4–6 layers, including basal
cells) was lower than in natural skin and other models of
bilayered skin equivalents.43,45,51,52 The ultrastructure of the
synthetic construct showed a typical structure in the basal
stratum with the natural epidermis.53 However, electron
microscopy analysis revealed that the overall organization
RIVA ET AL.
of these layers, as well as keratinocyte differentiation, was
poorer than with natural epidermis, including the formation
of junctional complexes, basement membrane, and keratinization.54 It may be hypothesized that low concentration
of calcium in the medium may partly account for incomplete development of desmosomal contacts and the cornified
envelope.43 Furthermore, it is presumable that bringing
the epidermal constructs to the air–liquid interface might
have improved keratinocyte stratification and keratinization
in our cultures, according to previous studies.11,55,56
Although no specific immunostaining for basement membrane proteins was observed underneath epithelial cells of
the constructs, laminin immunostaining was detectable
within the cytoplasm of basal cells. It may be that epithelial
cells are able to synthesize, but not secrete, laminin in the
dermal–epithelial interface. Our findings are in accordance
with previous studies showing that the formation of basement membrane at the dermal–epidermal junction, as well
as the synthesis of other matrices by dermal fibroblasts,
requires the interactions between keratinocytes and dermal
fibroblasts.43–45,57
Although the basement membrane was absent in our
constructs, basal keratinocytes expressed integrin b1 subunit and a2b1 integrin. Integrins are heterodimeric cellsurface glycoproteins that primarily mediate the attachment
of basal keratinocyte to extracellular matrix proteins.58–60 It
can be reasonably presumed that integrin expression is related to the presence of type I collagen that has been used
as a scaffold for epithelial growth in vitro.61,62 It has been
shown that a2b1 integrin is required for keratinocyte adhesion to type I and type IV collagens.63 The precise
mechanisms by which extracellular matrices influence cell
differentiation are largely unknown (for reviews see,64–66).
Accordingly, the molecular mechanisms regulating keratinocyte stratification by collagen are unclear. It is known
that cell-surface receptors for extracellular matrices (e.g.,
integrins) are involved in signaling pathways that activate
downstream proteins and genes within the cell.59,63,67 It
can be tentatively presumed that collagen is sufficient to
induce integrin expression by basal keratinocytes, which in
turn promotes keratinocyte differentiation and stratification. However, our experiments suggest that collagen alone
is not sufficient to support complete epidermal histogenesis,
according to the incomplete development of our artificial
constructs.
As indicated according to bromodeoxyuridine incorporation, cell proliferation occurs in the basal layer of the epithelial constructs, according to natural epidermis. Moreover,
the expression of p63, a nuclear factor involved in early
stages of epithelial differentiation, was detectable only in the
basal and suprabasal cell layers of the constructs, according
to the distribution observed in natural epidermis.68,69 Furthermore, although complete epithelial–mesenchymal interactions did not occur in the generation of our epidermal
constructs, differentiated keratinocytes of the upper layers
of the constructs were immunoreactive for filaggrin and
EPIDERMAL CONSTRUCT ON A COLLAGEN LAYER
9
involucrin, which are considered to be reliable markers of
keratinocyte terminal differentiation.70–72 At variance with
other models of TES,2–15 it is interesting to observe that
keratinocytes in our constructs underwent terminal differentiation at the molecular level without any interaction with
connective tissue cells in vitro, without increasing calcium
concentration in the culture medium,40 and without adjusting the constructs at the air–liquid interface.11,55
Our results demonstrate that human keratinocytes are
able to grow and differentiate on a collagen layer alone,
without the use of any feeder layer. This aspect may be
interesting for the production of synthetic epidermal constructs that are designed for clinical applications in which
the introduction of xenogenic cells or antigens should be
avoided. According to other models of TES, the constructs
that we generated may be useful in studying biological
interactions between epithelial and mesenchymal tissues. In
particular, this model may be used to explore which factors
are missing that the dermal layer would normally provide,
including decellularized matrices and soluble factors.
7. Sugihara, H., Toda, S., Miyabara, S., Kusaba, Y., and Minami,
Y. Reconstruction of the skin in three-dimensional collagen
gel matrix culture. In Vitro Cell Dev Biol 27, 142, 1991.
8. Andriani, F., Garfield, J., Fusenig, N.E., and Garlick, J.A.
Basement membrane proteins promote progression of intraepithelial neoplasia in 3-dimensional models of human stratified
epithelium. Int J Cancer 108, 348, 2004.
9. Kao, B., Kelly, K.M., Majaron, B., and Nelson, J.S. Novel
model for evaluation of epidermal preservation and dermal
collagen remodelling following photorejuvenation of human
skin. Laser Surg Med 32, 115, 2003.
10. Guerret, S., Govignoni, E., Hartmann, D.J., and Ronfard, V.
Long-term remodelling of a bilayered living human skin equivalent (Apligraf. grafted onto nude mice: immunolocalization
of human cells and characterization of extracellular matrix.
Wound Rep Reg 11, 35, 2003.
11. Asselineau, D., Bernard, B.A., Bailly, C., and Darmon, M.
Three-dimensional culture of human keratinocytes on a dermal equivalent as a model system to study environmental
modulation of epidermal physiology in vitro: Effect of airexposure. Br J Dermatol 31, 126, 1986.
12. Gallico, G.G. Biologic skin substitutes. Clin Plast Surg 17,
519, 1990.
13. Dubertret, L. Reconstruction of the human skin equivalent
in vitro: a new tool for skin biology. Skin Pharmacol 3, 144,
1990.
14. Garfein, E.V., Orgill, D.P., and Pribaz, J.J. Clinical applications
of tissue engineered constructs. Clin Plastic Surg 30, 485, 2003.
15. Falanga, V., Margolis, D., Alvarez, O., Auletta, M., Maggiacomo, F., Altrman, M., Jemsen, J., Sabolinski, M., HardinYoung, J., and the Human Skin Equivalent Investigators
Group. Rapid healing of venous ulcers and lack of clinical
rejection with an allogeneic cultured human skin equivalent.
Arch Dermatol 134, 293, 1998.
16. Boyce, S.T., and Ham, R.T. Calcium regulated of normal
human epidermal keratinocytes in chemically defined clonal
culture and serum-free serial culture. J Invest Dermatol 81,
33, 1983.
17. Carroll, J.M., and Moles, J.-P. A three dimensional skin
culture model for mouse keratinocytes: application to transgenic mouse keratinocytes. Exp Dermatol 9, 20, 2000.
18. Poumay, Y., Dupont, F., Marcoux, S., Leclercq-Smekens, M.,
Hérin, M., Coquette, A. A simple reconstructed human epidermis: preparation of the culture model and utilization in
in vitro studies. Arch Dermatol Res 296, 203, 2004.
19. Rosdy, M., Clauss, L.C. Terminal epidermal differentiation of
human keratinocyte grown in chemically defined medium on
inert filter substrates at the air-liquid interface. J Invest Dermatol 95, 409, 1990.
20. Falanga, V., Isaac, C., Paquette, D., Downin, G., Kouttab, N.,
Butmare, J., Badiavas, E., and Harding-Young, J. Wounding
of bioengineered skin: cellular and molecular aspects after
injury. J Invest Dermatol 119, 653, 2002.
21. Phillips, T.J. New skin for old; developments in biological
skin substitutes. Arch Dermatol 134, 344, 1998.
22. Eaglstein, W.H., and Falanga, V. Tissue engineering and the
development of Apligraf, a human skin equivalent. Clin Therap
19, 894, 1997.
23. De Luca, M., Franzi, A.T., D’Anna, F., Zicca, A., Albanese,
E., Bondanza, S., and Cancedda, R. Coculture of human
ACKNOWLEDGMENTS
We are grateful to Dr. Patrizia Vaghi for assistance with
laser scanning confocal microscopy analysis (Centro Grandi
Strumenti, University of Pavia) and Mrs. Aurora Farina and
Dr. Valeria Tinella (Department of Experimental Medicine,
University of Pavia) for assistance with immunohistochemistry. This research was supported by grants from the
University of Pavia (FAR) and Banca del Monte di Lombardia Foundation.
REFERENCES
1. Rheinwald, J.G., and Green, H. Serial cultivation of strains of
human epidermal keratinocytes: the formation of keratinizing
colonies from single cells. Cell 6, 331, 1975.
2. Auger, F.A., Lopez-Valle, C.A., Guignard, R., Tremblay, N.,
Noel, B., Goulet, F., and Germain, L. Skin equivalent produced with human collagen. In Vitro Cell Dev Biol 31, 432,
1995.
3. Parenteau, N.L., Nolte, C.M., Bilbo, P., Rosenberg, M.,
Wilkins, L.M., Johnson, E.W., Watson, S., Mason, V.S., and
Bell, E. Epidermis generated in vitro: practical considerations
and applications. J Cell Biochem 45, 245, 1991.
4. Bell, E., Ehrlich, H.P., Buttle, D.J., and Nakatsuji, T. Living
tissue formed in vitro and accepted as skin-equivalent tissue
of full thickness. Science 211, 1052, 1981.
5. Ateh, D.D., Vadgama, P., and Navsaria, H.A. Culture of human keratinocytes on polypyrrole-based conducting polymers.
Tissue Eng 12, 645, 2006.
6. El Ghalbzouri, A., Van den Bogaerdt, A.J., Kempenaar, J.,
and Ponec, M. Human adipose tissue derived cells delay reepithelialisation compared to skin fibroblasts in organotypic
skin culture. Brit J Dermatol 150, 444, 2004.
RIVA ET AL.
10
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
keratinocytes and melanocytes: differentiated melanocytes
are physiologically organized in the basal layer of the cultured
epithelium. Eur J Cell Biol 46, 176, 1988.
Archambault, M., Yaar, M., and Gilchrest B.A. Keratinocytes
and fibroblasts in a human skin equivalent model enhance
melanocyte and melanin synthesis after ultraviolet irradiation.
J Invest Dermatol 104, 859, 1995.
Regnier, M., Staquet, M.J., Schmitt, D., and Schmidt, R. Integration of Langerhans cells into a pigmented reconstructed
human epidermis. J Invest Dermatol 109, 510, 1997.
Black, A.F., Berthod, F., L’Heureux, N., Germain, L., and
Auger, F.A. In vitro reconstruction of a human capillary-like
network in a tissue-engineered skin equivalent. FASEB J 12,
1331, 1998.
Michel, M., L’Heureux, N., Pouliot, R., Xu, W., Auger, F.A.,
and Germain, L. Characterization of a new tissue-engineered
human skin equivalent with hair. In Vitro Cell Dev Biol Anim
35, 318, 1999.
Stark H.-J., Baur, M., Breitkreutz, D., Mirancea, N., and
Fusenig, N.E. Organotypic keratinocytes cocultures in defined
medium with regular epidermal morphogenesis and differentiation. J Invest Dermatol 112, 681, 1999.
Zacchi, V., Soranzo, C., Cortivo, R., Radice, M., Brun, P., and
Abatangelo, G. In vitro engineering of human skin-like tissue.
J Biomed Mater Res 40, 187, 1997.
Leigh, I.M., and Watt, F.M. Keratinocytes Methods. Cambridge, UK: Cambridge University Press, 1994.
Fujisaki, H., and Hattori, S. Keratinocytes apoptosis on type I
collagen gel caused by lack of laminin 5/10/11 deposition and
akt signalling. Exp Cell Res 280, 255, 2002.
Turksen, K., and Troy, T.C. Overexpression of the calcium
sensing receptor accelerates epidermal differentiation and
permeability barrier formation in vivo. Mech Dev 120, 733,
2003.
Yuspa, S.H., Kilkenny, A.E., Steinert, P.M., and Roop, D.R.
Expression of murine epidermal differentiation markers is
tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109, 1207, 1989.
Häkkinen, L., Koivisto, L., and Larjava, H. An improved
method for culture of epidermal keratinocytes from newborn
mouse skin. Meth Cell Sci 23, 189, 2002.
Minghetti, P., Casiraghi, A., Cilurzo, F., Montanari, L.,
Marazzi, M., Falcone, L., and Donati, V. Comparison of
different membranes with cultures of keratinocytes from man
for percutaneous absorption of nitroglycerine. J Pharmacol
51, 673, 1999.
Shi, S.R., Key, M.E., and Kalra, K.L. Antigen retrieval in
formalin-fixed, paraffin-embedded tissues: an enhancement
method for immunohistochemical staining based on microwave oven heating of tissue section. J Histochem Cytochem
39, 741, 1991.
Lavker, R.M., and Sun, T.T. Epidermal stem cells. J Invest.
Dermatol 81, 121, 1983.
Higham, M.C., Dawson, R., Szabo, M., Short, R., Haddow,
D.B., and MacNeil, S. Development of a stable chemically
defined surface for the culture of human keratinocytes under
serum free conditions for clinical use. Tissue Eng 9, 915,
2003.
Sun, T., Higham, M.C., Layton, C., Haycock, J.W., Short, R.,
and MacNeil, S. Development in xenobiotic-free culture of
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
human keratinocytes for clinical use. Wound Rep Reg 12,
626, 2004.
Komuves, L., Oda, Y., Tu, C.-L., Chang, W.H., Ho-Pao, C.L.,
Mauro, T., and Bikle, D.D. Epidermal expression of the fulllength extracellular calcium-sensing receptor is required for
normal keratinocytes differentiation. J Cell Physiol 192, 45,
2002.
Barrandon, Y., and Green, H. Three clonal types of keratinocytes with different capacities for multiplication. Proc Natl
Acad Sci USA 84, 2302, 1987.
Watt, F.M. The epidermal keratinocytes. Bioessays 8, 163,
1988.
Parenteau, N.L., Bilbo, P., Nolte, C.J., Mason, V.S., and
Rosenberg, M. The organotypic culture of human skin keratinocytes and fibroblasts to achieve form and function. Cytotechnology 9, 163, 1992.
Smola, H., Thiekotter, G., and Fusenig, N.E. Mutual induction of growth factor gene expression by epidermal-dermal
cell interactions. J Cell Biol 122, 417, 1993.
Casasco, A., Casasco, M., Zerbinati, N., Icaro Cornaglia, A.,
and Calligaro, A. Cell proliferation and differentiation in
a model of human skin equivalent. Anat Rec 264, 261, 2001.
Fuchs, E. Epidermal differentiation: the bare essentials. J Cell
Biol 111, 2807, 1990.
Parsa, R., Yang, A., McKeon, F., and Green, H. Association
of p63 with proliferative potential in normal and neoplastic
human keratinocytes. J Invest Dermatol 113, 1099, 1999.
Benard, J., Douc-Rasy S., and Ahomadegbe, J.C. TP53 family
members and human cancer. Hum Mutat 21, 182, 2003.
Janes, S.M., Lowell, S., and Hutter, C. Epidermal stem cell.
J Pathol 197, 479, 2002.
Moll, R., Franke, W.W., and Schiller, D.L. The catalog of
human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31, 11, 1982.
Nolte, C.J.M., Oleson, M.A., Bilbo, P.R., and Parenteau,
N.L. Development of stratum corneum and barrier function
in an organotypic skin culture. Arch Dermatol Res 285, 466,
1993.
Polak, J.M., and Van Noorden, S. Immunocytochemistry,
Modern Methods and Applications, 2nd Ed. Oxford: Butterworth Heinemann, 1986.
Zelickson, A.S. Ultrastructure of Normal and Abnormal Skin.
Philadelphia, PA: Lea & Febiger Press, 1967.
Fawcett, D.W., and Raviola, E. Skin. In: Bloom, W. and
Fawcett, D.W., eds. A Textbook of Histology, 12th Ed. New
York: Chapman & Hall, 1994, pp.525–558.
Harriger, M.D., and Hull, B.E. Cornification and basement
membrane formation in a bilayered human skin equivalent
maintained at an air-liquid interface. J Burn Care Rehabil 13,
187, 1992.
Sun, T., Norton, D., Haycock, J.W., Ryan, A.J., and MacNeil,
S. Development of a closed bioreactor system for culture of
tissue-engineered skin at an air-liquid interface. Tissue Eng
11, 1824, 2005.
El Ghalbzouri, A., Jonkman, M.F., Dijkman, R., and Ponec,
M. Basement membrane reconstruction in human skin equivalents is regulated by fibroblasts and/or exogenously activated
keratinocytes. J Invest Dermatol 124, 79, 2005.
Adams, J.C., and Watt, F.M. Changes in keratinocyte adhesion during terminal differentiation: reduction in fibronectin
EPIDERMAL CONSTRUCT ON A COLLAGEN LAYER
11
binding preceded a5B1 integrin loss from the cell surface.
Cell 63, 425, 1990.
Larjava, H, Peltonen, J., Akiyama, S.S., Gralnick, H.R., Uitto,
J., and Yamada, K.M. Novel function for beta 1 integrins in
keratinocyte cell-cell interactions. J Cell Biol 110, 803, 1990.
Hertle, M.D., Adams, J.C., and Watt, F.M. Integrin expression
during epidermal development in vivo and in vitro. Development 122, 193, 1991
Grinnel, F. Cell-collagen interactions: overview. Meth Enzymol 82, 499, 1982.
Peltonen, J., Larjava, H., Jaakkola, S., Grainick, H., Akiyama,
S.K., Yamada, S.S., Yamada, K.M., and Uitto, J. Localization
of integrin receptors for fibronectin, collagen, and laminin in
human skin. Variable expression in basal and squamous cell
carcinomas. J Clin Invest 84, 1916, 1989.
Zhang, Z.-G., Bothe, I., Hirche, F., Zweers, M., Gullberg, D.,
Pfitzer, G., Krieg, T., Eckes, B., and Aumalley, M. Interactions
of primary fibroblasts and keratinocytes with extracellular matrix proteins: contribution of alpha2beta1 integrin. J Cell Sci
119, 1886, 2006.
Ingber, D.E. Tensegrity II. How structural networks influence
cellular information processing networks. J Cell Sci 116,
1397, 2003.
Watt, F.M. Role of integrins in regulating epidermal adhesion, growth and differentiation. EMBO J 21, 3919, 2002.
Danen, E.H., and Sonnenberg, A. Integrins in regulation of
tissue development and function. J Pathol 201, 632, 2003.
Uings, I.J., and Farrow, S.N. Cell receptors and cell signalling. Mol Pathol 53, 295, 2000.
68. Koster, M.I., and Roop, D.R. The role of p63 in development
and differentiation of the epidermis. J Dermatol Sci 34, 3,
2004.
69. Pellegrini, G., Dellambra, E., Golisano, O., Martinelli, E.,
Fantozzi, I., Bondanza, S., Pozin, D., McKeon, F., and De
Luca, M. P63 identifies keratinocytes stem cells. Proc Natl
Acad Sci USA 98, 3156, 2001.
70. Galvin, S., Loomis, C., Manabe, M., Dhouailly, D., and Sun,
T.T. The major pathways of keratinocytes differentiation as
defined by keratin expression: an overview. Adv Dermatol 4,
277, 1989.
71. Morasso, M.I., and Tomic-Canic, M. Epidermal stem cell: the
cradle of epidermal determination, differentiation and wound
healing. Biol Cell 97, 173, 2005.
72. Gambardella, L., and Barrandon, Y. The multifaceted adult
epidermal stem cell. Curr Opin Cell Biol 15, 771, 2003.
59.
60.
61.
62.
63.
64.
65.
66.
67.
Address reprint requests to:
Federica Riva, Ph.D.
Department of Experimental Medicine
Histology and Embryology Unit
University of Pavia
Via Forlanini, 10
27100 Pavia
Italy
E-mail: [email protected]