RESULTS

Phenotypic characterization of unwounded human skin in the skin-humanized mice

Grafting of the fibrin-based skin construct regenerates in mice an EGFP-genetically modified human skin that closely resembles native human skin phenotype. The size and the permanence of the regenerated human skin allows for the induction of excisional full-thickness wounds. Details of the procedure are shown in Fig S1.

Although they can be easily distinguished by hematoxylin–eosin staining, the human and the murine skin were unequivocally identified by the expression of specific human markers and EGFP immunostaining (Fig S2a, b, and d). EGFP-modified human skin showed a well-differentiated and stratified epidermis with suprabasal layers immunopositive for human involucrin, denoting a normal development of the differentiation program (Fig S2a). Persistence of human dermal fibroblasts in the mature dermal area of the regenerated skin was evidenced by human vimentin expression (Fig S2b). Both human involucrin and vimentin staining depict the boundaries between mouse and human tissues.

In the unwounded skin, the epidermis overlay a mature collagenous matrix as revealed by Masson's trichrome staining (Fig S2c). The collagenous murine and human dermis appear as a continuous matrix with indistinguishable boundaries.

It is noteworthy that persistent EGFP expression in stable engrafted skin (Fig S2d) exerted a neutral effect on the physiology of transgenic human skin when compared with control (untransduced).

Wound healing in skin-humanized mice

The re-epithelialization process after wounding was monitored by analyzing data from the full extension of the wound. A comparative histological (H&E) analysis of the center of the wound on different days post-wounding was performed. Figure 3 shows how both the neoepidermis and the neodermis progressed toward the wound center after injury.

Figure 3.

Restoration of the differentiation program during wound repair in enhanced green fluorescent protein (EGFP)-bioengineered skin-humanized mice. Immunoperoxidase staining of loricrin. (a) Composite picture showing a panoramic view of a 1 d wound. (b) Higher magnification microphotograph taken from the right wound margin shown in panel a (rectangle). (c) Composite picture showing a panoramic view of a 3-d wound. Note the negative epithelium at the tip. (d) Higher magnification microphotograph taken from the middle zone of the right tongue shown in panel c (rectangle). (e). Composite picture showing a panoramic view of a 7-d wound. (f) Higher magnification microphotograph taken from the center of the wound shown in panel e (rectangle). Note loricrin was mostly expressed in the cytoplasm. (g) Composite picture showing a panoramic view at 15 d post-wounding. (h) Higher magnification microphotograph taken from the center of the wound shown in panel g (rectangle). Note restoration of the normal membranous pattern of loricrin expression in granular cells.

Full figure and legend (177K)

By 1 d post-wounding the wound margins (Figure 1a) were easily recognizable by an abrupt interruption in the epithelial (arrows) and dermal continuity. A scab covered the wound in close contact with the provisional matrix (asterisk).

Figure 1.

Re-epithelialization of wounds in skin-humanized mice. (a) Microphotograph taken at 1 d post-wounding. Wound margins indicated by arrows. (b) Microphotograph of a migratory epidermal tongue penetrating the wound bed (3 d post-wounding). (c) Microphotograph showing a re-epithelialized wound at 7 d post-wounding. In all panels, dermo-epidermal boundaries are outlined by dashed lines. Asterisks are placed on the temporary dermal matrix of the wound bed.

Full figure and legend (220K)

At 3 d post-wounding a migratory tongue of epidermal cells (outlined by a dashed line) penetrated into the wound bed (Figure 1b) under the scab. The forward part of the tongue (tongue tip) was composed of a monolayer, progressively followed by a bilayer and a disorganized, multilayered epithelium toward the wound margins (tongue tail).

At 7 d post-wounding, the center of the wound was fully covered by a stratified neoepithelium (Figure 1c; outlined by a dashed line). An immature neodermis (asterisk) underlay the neoepithelium. A similar picture was observed at 15 d post-wounding, although the neoepidermis lay on a more mature dermal bed (see Fig S4b vs c).

Species origin of the components involved in wound repair

The keratinocytes that re-epithelialized the wound expressed the EGFP protein, in all the different layers, as demonstrated 3 and 7 d post-wounding (completed coverage) (Fig S3a and b).

The use of specie specific antibodies (human involucrin and vimentin antisera) allowed us to confirm that the cells involved in the re-epithelialization and repopulation of the dermis after small excisional wounding were of human origin.

After 3 and 15 d, the wound bed was partially or completely invaded by keratinocytes that were immunopositive for human involucrin in the suprabasal layers (Fig S3c and d, respectively).

At 3 d post-wounding, no fibroblasts positive for human vimentin were present in the center of the wound bed. It is noteworthy that at this early time point of wound healing, dermal fibroblasts that were immunopositive for human vimentin, had just started dermal tissue remodelling (i.e., migration) and appeared far behind the epithelial tongue (Fig S3e). At 7 d post-wounding the wound bed was further populated by human vimentin immunopositive cells (Fig S3f).

Cytoskelet al plasticity changes during wound repair

The phenotypic changes of the epithelial migratory tongue were investigated in detail using several cytoskeleton markers. We have mainly focused on the changes occurring at 3 d post-wounding, a time point when wound closure is actively progressing (Figure 2).

Figure 2.

Cytoskeletal plasticity during wound repair in enhanced green fluorescent protein (EGFP)-bioengineered skin-humanized mice (3 d post-wounding). From panels a to e dermo-epidermal boundaries are outlined by dashed lines. (a) Immunoperoxidase staining of keratin K10 in a migratory epidermal tongue. Inset: normal pattern of keratin K10 expression in unwounded skin-humanized mouse. (b) Immunoperoxidase staining of keratin K16 in a migratory epidermal tongue. (c) DAPI nuclear staining of the migratory epidermal tongue shown in d and e. (d) Immunofluorescence staining of keratin K5 in the same section shown in panel c. Inset: normal pattern of keratin K5 expression in unwounded skin-humanized mouse. (e) Immunofluorescence staining of vimentin in the same section shown in panel c. Inset: normal pattern of vimentin expression in unwounded skin-humanized mouse. (f). Vimentin immunoperoxidase staining of a migratory epidermal tongue. (g). Close-up view of the marked area, black rectangle, in panel f.

Full figure and legend (131K)

One of the most prominent phenomena is the rearrangement of the intermediate filament cytoskeleton. Thus, keratins showed changes of expression toward the leading edge of the tongue. The expression of the suprabasal keratin K10 (Figure 2a, inset) progressively disappeared in the leading edge (Figure 2a) and the basal cell-specific keratin K5 (Figure 2d, inset) is also expressed in suprabasal keratinocytes (Figure 2c and d). Keratin K16 expression, a marker of keratinocyte hyperproliferation, which is highly induced at the wound margins gradually faded in the leading edge (Figure 2b). Remarkably, concomitant with the change in keratin markers, a robust vimentin expression in the keratinocytes of the epithelial tongue tip occurred (Figure 2c and e). Immunoperoxidase staining revealed, at high magnification, that some vimentin-positive keratinocytes with migratory phenotype appeared to roll-over other vimentin-positive keratinocytes along the tip of the tongue (Figure 2f and g).

Associated with the changes in keratin 5 and vimentin we have observed that the epithelial migratory tongue showed a marked decreased in E-cadherin immnunostaining toward the tongue tip (data not shown).

Dermal matrix regeneration

The kinetics of connective tissue regeneration was studied employing Masson's trichrome staining (Fig S4). The composition of the dermal matrix underwent changes from an original fibrin clot (1 d post-wounding) to a granulation tissue (3–7 d post-wounding) and a collagenous dermis (15 d post-wounding). At 1 d after injury, the wound was plugged by a provisional matrix containing blood vessels and invading cells consistent with inflammatory elements (Fig S4a and d). At 3 d post-wounding, a well-developed granulation tissue was established (data not shown). At 7 d after injury, the granulation tissue was rich in cells and still poor in collagen bundles as shown by Masson's trichrome light and diffuse blue staining (Fig S4b and e). At 15 d post-wounding, the density and labelling intensity of the collagen bundles increased whereas the number of cells decreased (Fig S4c). At this time the matrix still exhibited linear collagen bundles (Fig S4f). In contrast, in flanking areas of the wound, the connective tissue showed a higher intensity of Masson's trichrome staining (Fig S4g) and the collagen bundles acquired an "angel curl-shaped structure" typical of mature human skin.

Dermo-epidermal junction regeneration (DEJ)

As in native human skin, in the stable engrafted human skin human laminin was localized at the dermo-epidermal junction (Fig S5a).

As a result of the excisional wound through the entire thickness of the skin the basement membrane was removed. To assess for the reformation of basement membrane after wounding, expressions of two major components, i.e., type IV collagen and human laminin were monitored by immunostaining, using specific monoclonal antibodies. At 3 d after wounding, type IV collagen was absent at the distal tip of the migratory epidermal tongue (Fig S5b, outlined by a dashed line). Collagen type IV, however, became detectable underneath the newly stationary epidermis of the wound margin (Fig S5b, arrowhead). By 7 d, discontinuous labelling for laminin was observable in regions close to the wound center that was covered by a disorganized neoepithelium (Fig S5c, arrows). At 15 d post-wounding both collagen IV and laminin were present at the center of the wound (Fig S5d and data not shown, respectively).

Restoration of the differentiation program after wounding

To assess for the restoration of the late differentiation program, we studied the expression of loricrin, a marker of keratinocyte terminal differentiation. In the unwounded areas of the regenerated EGFP human skin, as in native human skin, expression of loricrin by keratinocytes was confined to the cornified and granular layers (Figure 3). By 1 d post-wounding (Figure 3a and b) a lack of loricrin expression was observed in the disorganized margin of the wound whereas in unwounded areas the expression was conserved. At 3 d post-wounding, the epithelial migrating tongue (Figure 3c and d) was negative for loricrin. In both cases, the unlabelled keratinocytes had an undifferentiated phenotype. Further from the tip of the tongue, toward the unwounded area, the expression of loricrin became progressively closer to normal. Seven days post-wounding (Figure 3e and f) a faint and discontinuous expression of loricrin was already present in the center of the wound. At 15 d post-wounding (Figure 3g and h) the expression of loricrin by the whole neoepithelium that covered the wound was completely restored.

Proliferative activity in response to wound

Restoration of normal skin architecture during wound healing results from both migration and mitosis. The kinetics of proliferative activity was monitored by evaluating bromodeoxyuridine (BrdU) incorporation in both the unwounded regenerated EGFP-human skin and at 1, 3, 7, and 15 d post-wounding (Fig S6). In unwounded regenerated EGFP-human skin cell proliferation occurred in epidermal and dermal compartments. Both keratinocytes and fibroblasts incorporated BrdU into the nuclei (Fig S6a, arrows). Labelling was restricted to a basal location in the epidermis. It is noteworthy that persistent EGFP expression in stable engrafted skin exerted a neutral effect on the BrdU labelling index of transgenic human skin when compared with untransduced control (3.5%0.6%vs 3.0%2.7% of basal keratinocytes; Fig S6d). A transient burst of proliferative activity occurred 1–3 d after wounding (24.5%0.4% and 28.4%16.4% of the basal epidermal cells, respectively; Fig S6b and d). Note that at 1 d post-wounding, the proliferative cells are accumulated in the tail of the tongue whereas at 3 d the accumulation tends to occur in the middle zone (Fig S6e). No keratinocyte incorporating BrdU, however, was observed at the tip of the migratory tongue (Fig S6b and e). At 7 d, the center of the wound still showed a marked proliferating activity that sometimes also involved suprabasal keratinocytes (13.1%3.6%; Fig S6d). At 15 d after wounding, the proliferative activity remained increased in the neoepithelium (9.4%1.3%; Fig S6d) when compared with unwounded areas.

rKGF improves wound healing in the skin-humanized mouse model

To determine the ability of the regenerated skin to respond to stimulatory agents, exogenous recombinant KGF was applied on healing wounds. The effect of KGF on the healing of 2 mm wounds was evaluated at 3 d post-wounding (third day after the beginning of treatment). Four criteria, namely, re-epithelialization, epidermal thickness, basal cell density, and proliferation index were used to quantify the effects of rKGF on the healing of stable engrafted human skin (Figure 4 and Table S1).

Figure 4.

rKGF improves wound healing in the skin-humanized mouse model. (a) Composite picture showing the center of a PBS-injected 3-d wound stained with H&E. (b) Composite picture showing the center of a KGF-injected 3-d wound stained with H&E. Note that rKGF protein reduces the epidermal gap distance. (c) Higher magnification of a PBS-injected skin-humanized mice. (d) Higher magnification of a KGF-injected skin-humanized mice. Note that rKGF protein increases epidermal thickness and basal cell density. (e) Immunoperoxidase bromodeoxyuridine (BrdU) staining of a PBS-injected 3-d wound. (f). Immunoperoxidase BrdU staining of a KGF-treated 3-d wound. Note that rKGF protein increases the number of BrdU-positive nuclei.

Full figure and legend (148K)

The percentage of re-epithelialization increased in KGF-injected wounds (Figure 4a vs b and Table S1). In addition, rKGF injection induced significant epidermal thickening (Figure 4c vs d and Table S1). Moreover, the evaluation of the number of basal cells per unit length of basement membrane showed that the density of basal keratinocytes was significantly higher in the rKGF-injected wounds (Figure 4a vs b and Table S1), and the cells were tightly packed and elongated. Finally, the proliferative response was significantly increased in KGF-injected wounds (1.5-fold increase, p<0.05; Figure 4e vs f). In keeping with previous findings for 3 d after wounding, the burst of proliferative activity occurred in the middle of the migratory tongue and no keratinocyte incorporating BrdU was observed at the tip of the migratory tongue (Figure 4e and f).

DISCUSSION

Previous wound-healing studies using genetically modified human skin grafted on mice have been focused on early time point analysis of graft behavior. The central aim of such studies was to examine the transient improvement of the graft take process. Thus, within 7–15 d post-grafting (a time frame that coincides with acute healing of the graft), the effects of the transferred genes were tested on the acceleration of graft invasion by fibrovascular cells and the deposition of collagen (^{Eming et al, 1995;Eming et al, 1998}, PDGF-modified grafts), on graft vascularization (^{Del Rio et al, 1999;Supp et al, 2000a, b;Supp and Boyce, 2002}, VEGF-modified grafts) and on epidermal proliferation (^{Hamoen and Morgan, 2002}, HGF-modified grafts). In contrast, the aim of our study was not to test short-term effects on grafting but rather to assess the wound-healing process of a small full-thickness wound in a mature, "quiescent" regenerated human skin that permanently expresses a transgene. Therefore, while previous studies refer to animal models that mimic the healing of a big-size wound that requires graft transplantation to achieve skin repair, such as grafting of a burn victim with a skin substitute, our model mimics small excisional wounds in volunteers.

One critical issue in our model is the persistence of transgene expression in the skin invivo. To the best of our knowledge, there is no other model system that allows for wound-healing studies in permanently modified human skin. Permanent gene expression occurs in our pre-clinical system through epidermal stem-cell targeting, as shown in two previous studies by our group (^{Del Rio et al, 2002;Serrano et al, 2003}). In keeping with these findings, in the present study, stable EGFP expression is consistent with targeting of the epidermal stem cell compartment since the length of expression spans multiple epidermal turnover cycles for regenerated human skin (3–4 wk duration per cycle;^{Robbins et al, 2001}). Experimental evidence indicates that fibrin, the material used as a dermal scaffold in our cultured skin equivalent, greatly perform to maintain the stem cell population invivo (^{Del Rio et al, 2002;Serrano et al, 2003}). Permanent epidermal regeneration has been shown in clinical practice when fibrin-based skin equivalents were grafted to severe burn patients, providing unequivocal evidence of functional stem cell persistence also in humans (^{Pellegrini et al, 1999;Ronfard et al, 2000;Llames et al, 2004}). Herein, we showed that persistent EGFP gene expression in the keratinocytes exerts a neutral effect on the physiology of the wound-healing process, allowing for wound healing studies in gene-transferred human skin after stable engraftment. Previously, in vitro studies of wound healing using human keratinocytes genetically marked with retroviral vectors have shown that targeted cells respond similarly to non-transduced cells (^{Garlick et al, 1991;Garlick and Taichman, 1994}).

Earlier studies in human subjects and in a diversity of animals (^{Odland and Ross, 1968;Croft and Tarin, 1970;Krawczyk, 1971;Winstanley, 1975a, b;Ortonne et al, 1981;Mansbridge and Knapp, 1987;Cavani et al, 1993}) showed that the healing of a cutaneous wound proceeds according to the following sequence of overlapping events: inflammation, production of granulation tissue, re-epithelialization (reconstruction of a DEJ), and remodelling of that tissue to form a neodermis (^{Clark, 1996}).

In the present study we provide invivo evidence that the genetically modified skin-humanized mouse model recapitulates the main features of native human wound healing, which we have carefully dissected using both epithelial and stromal markers.

Intermediate filament cytoskeleton plasticity appears as one of the hallmarks of the re-epithelialization event. Thus, K5 and K10, members of the major keratin pairs in the epidermis (^{Fuchs, 1990}), showed changes in their expression toward the wound tongue tip. Therefore, in the tongue tip K5 appears to be expressed in keratinocytes in various epidermal layers (including suprabasal layers) whereas expression of K10 gradually fades toward the tip. Concomitantly, expression of vimentin in basal keratinocytes occurs, indicating an epithelial–mesenchymal transition, a phenomenon that correlates with the acquisition of an active cell migratory phenotype (^{Steinert and Roop, 1988;Gibbins et al, 1999;Gilles et al, 1999;Boyer et al, 2000;Kiemer et al, 2001}). The transient focal co-expression of vimentin and cytokeratins by leading keratinocytes of the migratory tongue during epithelial wound healing has been previously reported in in vitro and exvivo studies (^{SundarRaj et al, 1992;Biddle and Spandau, 1996;Buisson et al, 1996}).

As predicted, in wounded transgenic human skin, K16 expression was induced after injury and was located in suprabasal keratinocytes at the wound edge, behind the migratory keratinocytes of the epidermal tongue, an area that presents a high BrdU labelling index. This result is consistent with previous data demonstrating K16 induction associated to hyperproliferation (^{Weiss et al, 1984;Leigh et al, 1995}) and involved in enabling the differentiating keratinocyte to become competent for re-epithelialization (^{Mansbridge and Knapp, 1987;Paladini et al, 1996}).

As the migratory epithelial tongue advanced from the surrounding margins, the expression of differentiation markers K10, and loricrin reproduced the centripetally progressive differentiation that occurs during re-epithelialization (^{Odland and Ross, 1968;Ortonne et al, 1981;Demarchez et al, 1986;Garlick and Taichman, 1994;Laplante et al, 2001}). In keeping with other studies on human wound healing, genetically modified human skin re-epithelialization was completed 7 d post-wounding (^{Demarchez et al, 1986;Cavani et al, 1993}). In the same way, the normal differentiation pattern, involving involucrin, K10 and loricrin, was restored in the transgenic skin-humanized mice 15 d after wounding (^{Odland and Ross, 1968;Ortonne et al, 1981;Demarchez et al, 1986;Garlick and Taichman, 1994;Laplante et al, 2001}).

During wound healing of gene-targeted skin-humanized mice, the re-epithelialization of the wound occurs by migration of human keratinocytes in contact with a chimearic human/mouse granulation tissue. We found that the sequence of events involving type IV collagen and laminin restoration is very similar to that described for wound healing in human volunteers (^{Odland and Ross, 1968;Ross and Odland, 1968;Larjava et al, 1993;Paladini et al, 1996;Burgeson and Christiano, 1997}).

During the early stages corresponding to the epidermal and basement membrane reconstruction phase, the granulation tissue (mostly of mouse origin) was rich in cells embedded in a loose matrix that was poor in collagen and highly vascularized. At 3 d post-wounding a migratory tongue of human fibroblasts started to invade the wound bed of the transgenic skin. Interestingly, Demarchez and collaborators working with wounds performed in unmodified human skin biopsies grafted onto nude mice showed that when human fibroblasts were present in the dermis of the graft, migration of host fibroblasts was avoided (^{Demarchez et al, 1987;Rossio-Pasquier et al, 1999}). Although the issue was not addressed in this study, it is very likely that collagen deposited in regions undergoing wound repair was of human origin.

As previously described in other in vitro and invivo wound-healing models (^{Matoltsy and Viziam, 1970;Winstanley, 1975b;Garlick and Taichman, 1994;Laplante et al, 2001;Falanga et al, 2002}), a burst of proliferative activity during re-epithelialization occurred in the skin-humanized mice. The burst occurred 1–3 d after wounding, during the process of covering of the wound floor. An elevated proliferative activity persisted around 7 d post-wounding, at the time that the epithelium covering the wound floor undergoes stratification. At this time labelled nuclei are also found in suprabasal layers revealing a faster turnover of the newly formed epithelium and a shorter time for cells to differentiate. These phenomena result in a thickened and hyperplastic neoepithelium as previously reported in animal models (^{Croft and Tarin, 1970;Krawczyk, 1971;Winstanley, 1975a, b}) and in human volunteers (^{Odland and Ross, 1968}). Furthermore, our results support previous studies demonstrating a delayed and transient increase in proliferative activity at the wound margin beginning after migration has initiated (^{Winstanley, 1975b;Garlick and Taichman, 1994;Laplante et al, 2001}).

Using this model, we have been able to capture certain images such as those of cells "rolling over one another", an observation consistent with the "frog leap" model proposed by^{Krawczyk (1971)} and still debated (^{Garlick and Taichman, 1994;Paladini et al, 1996;Laplante et al, 2001}).

Finally, our system is able to respond to rKGF added exogenously to wounds (^{Werner, 1998;Werner and Grose, 2003}). As shown in another model (^{Staiano-Coico et al, 1993;Pierce et al, 1994;Werner, 1998;Andreadis et al, 2001}), wounds performed in skin-humanized mice and treated with rKGF displayed an increase in epidermal thickness and proliferation that resulted in accelerated wound closure.

In conclusion, the gene-targeted skin-humanized mouse model would be useful to identify relevant genes involved in cutaneous repair, providing a complementary or alternative tool to transgenic mouse skin wound healing.

MATERIALS AND METHODS

Primary cultures of human keratinocytes and fibroblasts

Human keratinocytes and dermal fibroblasts were obtained from skin biopsies of donors by enzymatic digestion (^{Rheinwald and Green, 1975}). Primary keratinocytes were cultured on a feeder layer of lethally irradiated (X-ray; 50 Gy) 3T3-J2 cells (a gift from Dr J. Garlick, SUNY) as previously described (^{Meana et al, 1998;Del Rio et al, 2002}). The keratinocyte seeding media was a 3:1 mixture of Dulbecco's Modified Eagle Medium (DMEM) (GIBCO-BRL, Barcelona, Spain) and HAM'S F12 (GIBCO-BRL) containing 10% fetal calf serum (FCS), 0.1 nM choleric toxin, 2 nM T3, 5 g per mL insulin, and 0.4 g per mL hydrocortisone. Primary fibroblasts were cultured on plastic in DMEM containing 10% FCS. Cells were cultured at 37°C in a humid atmosphere containing 5% CO₂. The culture medium was changed every 2 d.

Gene transfer to primary keratinocytes

The packaging cell line PA317 (ATCC) was used to generate a stable cell line to produce amphotropic retroviral particles containing the pLZR-ires-EGFP sequence. Primary human keratinocytes were genetically modified by retroviral gene transfer according to^{Del Rio et al (2002)}. Keratinocytes were analyzed for EGFP expression and selected by fluorescence-activated cell sorting (FACS) on a FACStar PLUS flow cytometer (Becton Dickinson, San Jose, California).

Wound-healing experimental design in skin–humanized mice

The invivo wound-healing model was designed by wounding a stable engrafted human skin on nude mice (9–12 wk post grafting). The bioengineered human skin equivalent is based on the use of fibrin containing live fibroblasts as a dermal component (^{Meana et al, 1998}). Briefly, 3 mL of the fibrinogen (from cryoprecipitates) were added to 12 mL of DMEM with 10% FCS containing 5 10⁵ dermal fibroblasts and x500 IU of bovine aprotinin (Trasylol; Bayer, West Haven, Connecticut). Immediately afterwards, 1 mL of 0.025 mM CaCl₂ (Sigma, St Louis, Missouri) with 11 IU of bovine thrombin (Sigma) were added. Finally, the mixture was placed on a six-well culture plate (Transwell; Costar, Cambridge, Massachusetts) and allowed to solidify at 37°C for 2 h. Untransduced control or purified EGFP+keratinocytes obtained by cell sorting were then seeded and grown submerged up to keratinocyte confluence. Mice were aseptically cleansed and grafted as previously described elsewhere (^{Del Rio et al, 2002;Serrano et al, 2003;Llames et al, 2004}). Briefly, full thickness 12 mm circular wounds were then created on the dorsum of the mice. Untransduced control or EGFP-bioengineered equivalents were placed orthotopically on the wound. At 9–12 wk after transplantation, these stable engrafted human skins were injured with 2 mm biopsy punches (STIEFEL LAB, Madrid, Spain). At 1, 3, 5, 7, and 15 d post-wounding, mice were euthanized by CO₂ asphyxiation. Rectangular samples of skin containing the wounds in the center were harvested and fixed in 3.7% formaldehyde or 70% ethanol solution. After fixation the skin biopsy was embedded in paraffin. Serial 4 m cross sections were obtained. The whole sample was sectioned to determine the center of the wound and adequately monitor the healing process.

Mice were housed for the duration of the experiment at the CIEMAT Laboratory Animals Facility (Spanish registration number 28079-21 A) in pathogen-free conditions using individually ventilated type II cages (25 air changes per hour) and 10 KGy -irradiated soft wood pellets as bedding. All handling was carried out according to European and Spanish laws and regulations (European Convention 123, use and protection of Vertebrate Mammals in experimentation and other scientific purposes. Spanish R.D. 223/88 and O.M. 13-10-89 of the Ministry of Agricultural, Food and Fisheries, protection and use of animals in scientific research and internal biosafety and bioethics guidelines).

Analysis of the wound-healing process: techniques of histology and immunohistochemistry

The wound-healing process, either in untransduced control or EGFP-modified human skin, was analyzed by histological and immunohistochemical techniques. Paraffin sections were dewaxed by melting for 30–60 min at 60°C, cleared in xylene three times for 5 min, and re-hydrated in water solutions containing decreasing percentages of ethanol. To determine tissue architecture, sections were stained with hematoxylin–eosin (Gill 2 Haematoxylin and Eosin Y alcoholic; Thermo Sandon, Cheshire, UK) following a standard procedure.

Two methods were used to determine the EGFP reporter gene expression. Green fluorescence was readily visualized in the intact xenograft invivo using a fluorescence stereomicroscope under blue light (Olympus, Olympus America, Melville, New York). Formaldehyde fixed sections were used to perform immunohistochemistry using specific polyclonal antibodies anti-EGFP (clone A-11122, Molecular Probes, Eugene, Oregon) at a 1:200 dilution.

To describe the maturation program, formaldehyde (immunoperoxidase) or 70%-ethanol (immunofluorescence) fixed sections were stained using specific antibodies against human epidermal differentiation markers. The antibodies were used at final dilutions of 1:600 for keratin K5 polyclonal (clone AF138: BabCO, Richmond, California), of 1:50 for -keratin 10 monoclonal (clone AE2: ICN Biomedicals, Aurora, Ohio), of 1:3 for keratin 16 monoclonal (clone LL025 a gift from Dr Lane, Dundee University, UK), 1:100 for involucrin human-specific monoclonal antibody (clone SY5: Sigma) and 1:2000 for loricrin polyclonal (clone AF-62: BabCO, Ca). The presence of cell–cell adherent junctions of epithelial cells was monitored by the use of antibodies specific for e-cadherin (clone 36, Transduction Lab, Lexington, Kentucky).

To determine the regeneration of the dermo-epidermal junction, ethanol-fixed sections were used. In the case of immunofluorescence, sections were digested with 0.1% trypsin (ICN Biomedicals) for 30 min. The sections were immunostained with specific antibodies against two main components of the basement membrane: laminin (clone LAM-89 Sigma) and collagen IV (clone PHM-12+CIV 22: NeoMarkers, Fremont, California). Antibodies were used at a final dilution of 1:200.

To evaluate dermal remodelling, formaldehyde fixed sections were analyzed. Human fibroblasts were labelled using monoclonal antibody to human vimentin (V9: BioGenex, San Romon, California) at a final dilution of 1:50. The vimentin antibody does not react with mice tissue. Therefore it provides additional proof of the human origin of the regenerated dermis on the nude mice. Masson's trichrome (Acoustain Trichrome Stains-Masson; Sigma) was used as a marker of collagen fibres following manufacturer's recommendations.

In all cases, immunohistochemistry was performed using standard procedures. Immunoperoxidase staining was developed using the Vectastain ABC kit (Vector, Burlingame, California). The maximum diameter in open wounds was selected by light microscopy observation. The whole wound was sectioned and 1 of every 10 sections was stained with hematoxylin–eosin. In covered wounds, the more immature area of the wound was determined by Masson's trichrome staining. The scab was also a useful landmark in both cases.

Cell proliferation during wound healing

Grafted mice were IP injected with 20 mg per kg of BrdU 1 h before euthanasia. To assess proliferation, immunohistochemistry using specific antibodies against BrdU (Roche, Indianapolis) was performed. The BrdU primary antibody was diluted 1:50. The number of BrdU-positive basal nuclei in a unit length of 2.5 mm (containing the center of the wound in wounded samples) was counted. A minimum of three animals was used for each time point.

rKGF treatment of wounds

Twelve weeks post-grafting the regenerated human skin was wounded using 2 mm biopsy-punches as described above. Wounds were injected during three consecutive days with 1.5 g of recombinant human KGF in 100 L of PBS (n=3) or PBS alone (n=3). The total volume was administered via three separate injections around the wound margin. Samples were harvested 3 d post-wounding, formalin fixed and histologically processed. The percent of re-epithelialization across each wound site was measured by light microscopy using a reticle to measure the proportion of each wound that was covered by neoepidermis in relation to the entire wound length (^{Staiano-Coico et al, 1993;Pierce et al, 1994}). The re-epithelialization percentage was calculated by the formula: 100 [(wound diameter-epidermal gap)/wound diameter]. The epidermal gap is the distance between opposite epithelial tongues.

Epidermal thickness was defined as the distance (in millimeters) from the top of the granular layer to the bottom of the basal layer (^{Staiano-Coico et al, 1993}). Light microscopy using an ocular reticle was used for these measurements. Approximately 30 individual measurements were made along the wound margin for each histologic section and the mean thickness was evaluated.

Basal cell density was defined as the number of basal cells per mm of the basement membrane (^{Andreadis et al, 2001}). Finally, the percentage of BrdU-positive basal nuclei was calculated by counting 400 cells at the wound site.

Statistics

A Student's t test was applied to compare the means of samples. Differences were considered statistically significant when p<0.05.

References

1. Andreadis, ST, Hamoen, KE, Yarmush, ML, Morgan, JR: Keratinocyte growth factor induces hyperproliferation and delays differentiation in a skin equivalent model system. FASEB J 2001 15:898–906, 10.1096/fj.00-0324com | Article | PubMed | ISI | ChemPort |
2. Bereiter-Hahn, J: Epidermal cell migration and wound repair. In: Bereiter-Hahn J, Matoltsy AG, Richards KS (eds). Biology of the Integuments. Vol. 2: Vertebrates 1986 Berlin: Springer-Verlag, p 443–471,
3. Biddle, D, Spandau, DF: Expression of vimentin in cultured human keratinocytes is associated with cell–extracellular matrix junctions. Arch Dermatol Res 1996 288:621–624,  | Article | PubMed | ISI | ChemPort |
4. Boyer, B, Valles, AM, Edme, N: Induction and regulation of epithelial–mesenchymal transitions. Biochem Pharmacol 2000 60:1091–1099,  | Article | PubMed | ISI | ChemPort |
5. Buisson, AC, Gilles, C, Polette, M, Zahm, JM, Birembaut, P, Tournier, JM: Wound repair-induced expression of a stromelysins is associated with the acquisition of a mesenchymal phenotype in human respiratory epithelial cells. Lab Invest 1996 74:658–669,  | PubMed | ISI | ChemPort |
6. Burgeson, RE, Christiano, AM: The dermal–epidermal junction. Curr Opin Cell Biol 1997 9:651–658,  | Article | PubMed | ISI | ChemPort |
7. Cavani, A, Zambruno, G, Marconi, A, Manca, V, Marchetti, M, Giannetti, A: Distinctive integrin expression in the newly forming epidermis during wound healing in humans. J Invest Dermatol 1993 101:600–604,  | Article | PubMed | ISI | ChemPort |
8. Clark, RAF: The molecular and cellular biology of wound repair. 1996 New York and London: Plenum Press,
9. Croft, CB, Tarin, D: Ultrastructural studies of wound healing in mouse skin. I. Epithelial behaviour. J Anat 1970 106:63–77,  | PubMed | ISI | ChemPort |
10. Del Rio, M, Larcher, F, Meana, A, Segovia, J, Alvarez, A, Jorcano, J: Nonviral transfer of genes to pig primary keratinocytes. Induction of angiogenesis by composite grafts of modified keratinocytes overexpressing VEGF driven by a keratin promoter. Gene Ther 1999 6:1734–1741, 10.1038/sj.gt.3300986 | Article | PubMed | ISI | ChemPort |
11. Del Rio, M, Larcher, F, Serrano, F, et al: A preclinical model for the analysis of genetically modified human skin invivo. Hum Gene Ther 2002 13:959–968, 10.1089/10430340252939069 | Article | PubMed | ISI | ChemPort |
12. Demarchez, M, Hartmann, DJ, Herbage, D, Ville, G, Prunieras, M: Wound healing of human skin transplanted onto the nude mouse. II. An immunohistological and ultrastructural study of the epidermal basement membrane zone reconstruction and connective tissue reorganization. Dev Biol 1987 121:119–129, 10.1016/0012-1606(87)90145-X | Article | PubMed | ISI | ChemPort |
13. Demarchez, M, Sengel, P, Prunieras, M: Wound healing of human skin transplanted onto the nude mouse. I. An immunohistological study of the re-epithelialization process. Dev Biol 1986 113:90–96, 10.1016/0012-1606(86)90110-7 | Article | PubMed | ISI | ChemPort |
14. Eming, SA, Lee, J, Snow, RG, Tompkins, RG, Yarmush, ML, Morgan, JR: Genetically modified human epidermis overexpressing PDGF-A directs the development of a cellular and vascular connective tissue stroma when transplanted to athymic mice—Implications for the use of genetically modified keratinocytes to modulate dermal regeneration. J Invest Dermatol 1995 105:756–763, 10.1111/1523-1747.ep12325550 | Article | PubMed | ISI | ChemPort |
15. Eming, SA, Medalie, DA, Tompkins, RG, Yarmush, ML, Morgan, JR: Genetically modified human keratinocytes overexpressing PDGF-A enhance the performance of a composite skin graft. Hum Gene Ther 1998 9:529–539,  | PubMed | ISI | ChemPort |
16. Falanga, V, Isaacs, C, Paquette, D, et al: Wounding of bioengineered skin: Cellular and molecular aspects after injury. J Invest Dermatol 2002 119:653–660, 10.1046/j.1523-1747.2002.01865.x | Article | PubMed | ISI | ChemPort |
17. Fuchs, E: Epidermal differentiation: The bare essentials. J Cell Biol 1990 111:2807–2814, 10.1083/jcb.111.6.2807 | Article | PubMed | ISI | ChemPort |
18. Garlick, JA, Katz, AB, Fenjves, ES, Taichman, LB: Retrovirus-mediated transduction of cultured epidermal keratinocytes. J Invest Dermatol 1991 97:824–829, 10.1111/1523-1747.ep12489019 | Article | PubMed | ISI | ChemPort |
19. Garlick, JA, Taichman, LB: Fate of human keratinocytes during re-epithelialization in an organotypic culture model. Lab Invest 1994 70:916–924,  | PubMed | ISI | ChemPort |
20. Gibbins, JR, Manthey, A, Tazawa, YM, Scott, B, Bloch-Zupan, A, Hunter, N: Midline fusion in the formation of the secondary palate anticipated by upregulation of keratin K5/6 and localized expression of vimentin mRNA in medial edge epithelium. Int J Dev Biol 1999 43:237–244,  | PubMed | ISI | ChemPort |
21. Gilles, C, Polette, M, Zahm, JM, Tournier, JM, Volders, L, Foidart, JM, Birembaut, P: Vimentin contributes to human mammary epithelial cell migration. J Cell Sci 1999 112:4615–4325,  | PubMed | ISI | ChemPort |
22. Hamoen, KE, Morgan, JR: Transient hyperproliferation of a transgenic human epidermis expressing hepatocyte growth factor. Cell Transplant 2002 11:385–395,  | PubMed | ISI |
23. Kiemer, AK, Takeuchi, K, Quinlan, MP: Identification of genes involved in epithelial-mesenchymal transition and tumor progression. Oncogene 2001 20:6679–6688, 10.1038/sj.onc.1204872 | Article | PubMed | ISI | ChemPort |
24. Krawczyk, WS: A pattern of epidermal cell migration during wound healing. J Cell Biol 1971 49:247–263, 10.1083/jcb.49.2.247 | Article | ISI |
25. Laplante, AF, Germain, L, Auger, FA, Moulin, V: Mechanisms of wound re-epithelialization: Hints from a tissue-engineered reconstructed skin to long-standing questions. FASEB J 2001 15:2377–2389, 10.1096/fj.01-0250com | Article | PubMed | ISI | ChemPort |
26. Larcher, F, Del Rio, M, Serrano, F, et al: A cutaneous gene therapy approach to human leptin deficiencies: Correction of the murine ob/ob phenotype using leptin-targeted keratinocyte grafts. Faseb J 2001 15:1529–1538, 10.1096/fj.01-0082com | Article | PubMed | ISI | ChemPort |
27. Larjava, H, Salo, T, Haapasalmi, K, Kramer, RH, Heino, J: Expression of integrins and basement membrane components by wound keratinocytes. J Clin Invest 1993 92:1425–1435,  | PubMed | ISI | ChemPort |
28. Leigh, IM, Navsaria, H, Purkis, PE, McKay, IA, Bowden, PE, Riddle, PN: Keratins (K16 and K17) as markers of keratinocyte hyperproliferation in psoriasis invivo and in vitro. Br J Dermatol 1995 133:501–511,  | PubMed | ISI | ChemPort |
29. Llames, SG, Del Rio, M, Larcher, F, et al: Human plasma as a dermal scaffold for the generation of a completely autologous bioengineered skin. Transplantation 2004 77:350–355, 10.1097/01.TP.0000112381.80964.85 | Article | PubMed | ISI |
30. Mansbridge, JN, Knapp, AM: Changes in keratinocyte maturation during wound healing. J Invest Dermatol 1987 89:253–263, 10.1111/1523-1747.ep12471216 | Article | PubMed | ISI | ChemPort |
31. Martin, P: Wound healing-aiming for perfect skin regeneration. Science 1997 276:75–81, 10.1126/science.276.5309.75 | Article | PubMed | ISI | ChemPort |
32. Matoltsy, AG, Viziam, CB: Further observations on epithelialization of small wounds: An autoradiographic study of incorporation and distribution of 3H-thymidine in the epithelium covering skin wounds. J Invest Dermatol 1970 55:20–25, 10.1111/1523-1747.ep12290488 | Article | PubMed | ISI | ChemPort |
33. Meana, A, Iglesias, J, Del Rio, M, et al: Large surface of cultured human epithelium obtained on a dermal matrix based on live fibroblast-containing fibrin gels. Burns 1998 24:621–630,  | Article | PubMed | ISI | ChemPort |
34. Odland, G, Ross, R: Human wound repair. I. Epidermal regeneration. J Cell Biol 1968 39:135–151, 10.1083/jcb.39.1.135 | Article | PubMed | ISI | ChemPort |
35. Ortonne, JP, Loning, T, Schmitt, D, Thivolet, J: Immunomorphological and ultrastructural aspects of keratinocyte migration in epidermal wound healing. Virchows Arch A Pathol Anat Histol 1981 392:217–230, 10.1007/BF00430822 | Article | PubMed | ChemPort |
36. Paladini, RD, Takahash, K, Bravo, NS, Coulombe, PA: Onset of re-epithelialization after skin injury correlates with a reorganization of keratin filaments in wound edge keratinocytes: Defining a potential role for keratin 16. J Cell Biol 1996 132:381–397, 10.1083/jcb.132.3.381 | Article | PubMed | ISI | ChemPort |
37. Pellegrini, G, Ranno, R, Stracuzzi, G, et al: The control of epidermal stem cells (holoclones) in the treatment of massive full-thickness burns with autologous keratinocytes cultured on fibrin. Transplantation 1999 68:868–879, 10.1097/00007890-199909270-00021 | Article | PubMed | ISI | ChemPort |
38. Pierce, GF, Yanagihara, D, Klopchin, K, Danilenki, DM, Hsu, E, Kenney, WC, Morris, CF: Stimulation of all epithelial elements during skin regeneration by keratinocyte growth factor. J Exp Med 1994 179:831–840, 10.1084/jem.179.3.831 | Article | PubMed | ISI | ChemPort |
39. Rheinwald, JG, Green, H: Serial cultivation of strains of human epidermal keratinocytes: The formation of keratinizing colonies from single cells. Cell 1975 6:331–343,  | Article | PubMed | ISI | ChemPort |
40. Robbins, PB, Lin, Q, Goodnough, JB, Tian, H, Chen, X, Khavari, PA: In vivo restoration of laminin 5 beta 3 expression and function in junctional epidermolysis bullosa. Proc Natl Acad Sci USA 2001 98:5193–5198, 10.1073/pnas.091484998 | Article | PubMed | ChemPort |
41. Ronfard, V, Rives, JM, Neveux, Y, Carsin, H, Barrandon, Y: Long-term regeneration of human epidermis on third degree burns transplanted with autologous cultured epithelium grown on a fibrin matrix. Transplantation 2000 70:1588–1598, 10.1097/00007890-200012150-00009 | PubMed | ISI | ChemPort |
42. Ross, R, Odland, G: Human wound repair. II. Inflammatory cells, epithelial–mesenchymal interrelations, and fibrogenesis. J Cell Biol 1968 39:152–168, 10.1083/jcb.39.1.152 | Article | PubMed | ISI | ChemPort |
43. Rossio-Pasquier, P, Casanova, D, Jomard, A, Demarchez, M: Wound healing of human skin transplanted onto the nude mouse after a superficial excisional injury: Human dermal reconstruction is achieved in several steps by two different fibroblast subpopulations. Arch Dermatol Res 1999 291:591–599, 10.1007/s004030050460 | Article | PubMed | ISI | ChemPort |
44. Scheid, A, Meuli, M, Gassmann, M, Wenger, RH: Genetically modified mouse models in studies on cutaneous wound healing. Exp Physiol 2000 85:687–704, 10.1111/j.1469-445X.2000.02103.x | Article | PubMed | ISI | ChemPort |
45. Serrano, F, Del Rio, M, Larcher, F, et al: A comparison of targeting performance of oncoretroviral versus lentiviral vectors on human keratinocytes. Hum Gene Ther 2003 14:1579–1585, 10.1089/104303403322495089 | Article | PubMed | ISI | ChemPort |
46. Singer, AJ, Clark, RAF: Cutaneous wound healing. N Engl J Med 1999 341:738–746, 10.1056/NEJM199909023411006 | Article | PubMed | ISI | ChemPort |
47. Spirito, F, Meneguzzi, G, Danos, O, Mezzina, M: Cutaneous gene transfer and therapy: The present and the future. J Gene Med 2001 3:21–31, 10.1002/1521-2254(2000)9999:9999&#60;::AID-JGM156>3.0.CO;2-I | Article | PubMed | ISI | ChemPort |
48. Staiano-Coico, L, Krueger, JG, Rubin, JS, et al: Human keratinocyte growth factor effects in a porcine model of epidermal wound healing. J Exp Med 1993 178:865–878, 10.1084/jem.178.3.865 | Article | PubMed | ChemPort |
49. Steinert, PM, Roop, DR: Molecular and cellular biology of intermediate filaments. Annu Rev Biochem 1988 57:593–625, 10.1146/annurev.bi.57.070188.003113 | Article | PubMed | ISI | ChemPort |
50. SundarRaj, NJ, Rizzo, D, Anderson, SC, Gesiotto, JP: Expression of vimentin by rabbit corneal epithelial cells during wound repair. Cell Tissue Res 1992 267:347–356, 10.1007/BF00302973 | Article | PubMed | ISI | ChemPort |
51. Supp, DM, Bell, SM, Morgan, JR, Boyce, ST: Genetic modification of cultured skin substitutes by transduction of human keratinocytes and fibroblasts with platelet-derived growth factor-A. Wound Repair Regen 2000a 8:26–35, 10.1046/j.1524-475x.2000.00026.x | Article | PubMed | ISI | ChemPort |
52. Supp, DM, Boyce, ST: Overexpression of vascular endothelial growth factor accelerates early vascularization and improves healing of genetically modified cultured skin substitutes. J Burn Care Rehabil 2002 23:10–20, 10.1097/00004630-200201000-00004 | Article | PubMed |
53. Supp, DM, Supp, AP, Bell, SM, Boyce, ST: Enhanced vascularization of cultured skin substitutes genetically modified to overexpress vascular endothelial growth factor. J Invest Dermatol 2000b 114:5–13, 10.1046/j.1523-1747.2000.00824.x | Article | PubMed | ISI | ChemPort |
54. Tarin, D, Croft, CB: Ultrastructural studies of wound healing in mouse skin. II. Dermo-epidermal interrelationships. J Anat 1970 106:79–91,  | PubMed | ISI | ChemPort |
55. Weiss, RA, Eichner, R, Sun, TT: Monoclonal antibody analysis of keratin expression in epidermal diseases: A 48- and 56-kdalton keratin as molecular markers for hyperproliferative keratinocytes. J Cell Biol 1984 98:1397–1406, 10.1083/jcb.98.4.1397 | Article | PubMed | ISI | ChemPort |
56. Werner, S: Keratinocyte growth factor: A unique player in epithelial repair processes. Cytokine Growth Factor Rev 1998 9:153–165, 10.1016/S1359-6101(98)00010-0 | Article | PubMed | ISI | ChemPort |
57. Werner, S, Breeden, M, Hubner, G, Greenhalgh, DG, Longaker, MT: Induction of keratinocyte growth factor expression is reduced and delayed during wound healing in the genetically diabetic mouse. J Invest Dermatol 1994 103:469–473, 10.1111/1523-1747.ep12395564 | Article | PubMed | ISI | ChemPort |
58. Werner, S, Grose, R: Regulation of wound healing by growth factors and cytokines. Physiol Rev 2003 83:835–870,  | PubMed | ISI | ChemPort |
59. Winstanley, EW: The epithelial reaction in the healing of excised cutaneous wounds in the dog. J Comp Pathol 1975a 85:61–75, 10.1016/0021-9975(75)90085-7 | Article | ISI | ChemPort |
60. Winstanley, EW: The rate of mitotic division in regenerating epithelium in the dog. Res Vet Sci 1975b 18:144–148,  | ISI | ChemPort |

Acknowledgments

We wish to thank Dr Alberto Alvarez and Israel Orman for FACS analysis, Almudena Holguín, Blanca Duarte, Isabel de los Santos, Pilar Hernández, and Sergio García for excellent technical assistance, Jesús Martínez for animal care and Soledad Moreno for help with artwork. We gratefully acknowledge the useful discussion and critical comments by Dr Marta Carretero, Ciemat. M.J.E. is a recipient of a postdoctoral fellowship from Pfizer/Fundación Marcelino Botín. M.G. is a recipient of a predoctoral fellowship from CAM. This work was supported by grants from the Spanish Ministry for Science and Technology (BMC 2001-1018), from the Spanish Ministry of Health (FIS 01/0556) and from the Community of Madrid (CAM 08.6/0004.1/2003).