Introduction

Type VII collagen, a non-fibrillar collagen, is a major component of anchoring fibril loop structures beneath the epidermal basement membrane (Uitto et al., 1992; Burgeson, 1993). Cloning of collagen VII cDNA demonstrated a primary sequence of 2,944 amino acids and the basic organization of the functional domains (Christiano et al., 1994a). Subsequent genomic cloning has highlighted the structural organization of the collagen VII gene (COL7A1) (Christiano et al., 1994b). This cloning information has enabled genomic DNA sequence analysis of COL7A1 and has demonstrated that mutations within COL7A1 are associated with the dystrophic forms of epidermolysis bullosa (DEB). DEB comprises a group of mechanobullous diseases characterized by cutaneous fragility with a tendency to form sub-basal lamina densa blisters (Christiano et al., 1993; Pulkkinen and Uitto, 1999; Chen et al., 2002a). In addition, targeted disruption of COL7A1 in a mouse model demonstrated an almost identical phenotype to DEB in humans (Heinonen et al., 1999). These results indicate that collagen VII is of critical importance for dermal–epidermal adhesion.

Approximately 300 distinct COL7A1 mutations have been identified in DEB patients so far. Therapeutic introduction of COL7A1 into skin cells is a promising treatment of DEB. Despite the relatively large size of COL7A1, the cDNA of which is still 9 kb makes gene transfer relatively problematic, and several methods including lentivirus- (Chen et al., 2002b), retrovirus- (Baldeschi et al., 2003) and C31 integrase-based approaches (Ortiz-Urda et al., 2002) have attempted to transfer COL7A1 into keratinocytes. These studies used keratinocytes as target cells as collagen VII has been reported to be mainly synthesized and secreted by keratinocytes and to lesser extent by fibroblasts (Ryynanen et al., 1992). However, application of gene-transferred DEB fibroblasts into the skin restored collagen VII expression in the dermal–epidermal junction (Ortiz-Urda et al., 2003; Woodley et al., 2003). In addition, using an intradermal injection of lentivirus with COL7A1 induced the expression of collagen VII in fibroblasts and endothelial cells, resulting in collagen VII accumulation in the grafted DEB skin on the host animal (Woodley et al., 2004).

In this study, we have established a retroviral method to transfer COL7A1 into DEB keratinocytes and fibroblasts. Next, we produced the skin grafts with gene-transferred keratinocytes or fibroblasts, and transplanted them into nude rats. Examination of collagen VII graft expression revealed that gene-transferred fibroblasts assembled more collagen VII in the form of anchoring fibrils beneath the basement membrane than gene-transferred keratinocytes. We conclude that fibroblasts are a more ideally suited target for COL7A1 gene transfer than keratinocytes using retroviral gene therapy for the treatment of DEB.

Results

Successful transfer COL7A1 using retroviral systems

We employed two retroviral vectors, pLIXN and pDON-AI, and full-length COL7A1 cDNA was inserted into the retroviral vectors to generate plasmids termed pLI-COL and pDON-COL, respectively (Figure 1). Also, we created pDON() by removing the SV-40 promoter and Neo gene from pDON-AI and constructed a retroviral vector with COL7A1 cDNA pDON()-COL (Figure 1). Several series of preliminary experiments demonstrated that retronectin (TAKARA) increased attachment of virus to keratinocytes and fibroblasts. Also, use of plasmid G protein of vesicular stomatitis virus (pVSV-G) (Pantropic System; Clontech, Palo Alto, CA) enabled concentration of viral particles by ultracentrifugation, resulting in an increase of transfer efficacy. After transfection of plasmids pLI-COL, pDON-COL, and pDON()-COL to 293 packaging cells, the culture media were collected. The virus titers (meanSD 10⁶/ml) of pLI-COL, pDON-COL, and pDON()-COL were 1.1+0.35, 1.6+0.46, and 2.7+0.55, respectively.

Figure 1.

Schematic representation of retroviral COL7A1 expression vectors. We employed two retroviral vectors, pLIXN (Clontech) and pDON-AI (Takara, Japan), and full-length COL7A1 cDNA was inserted into the retroviral vectors to generate plasmids pLI-COL and pDON-COL, respectively. A pDON() vector was created by removing the SV-40 promoter and Neo gene from pDON-AI and COL7A1 cDNA constructed retroviral vector made termed pDON()-COL. These vectors harbor long terminal repeat (LTR) derived from mouse moloney leukemia virus (MMLV) and human cytomegalovirus (HCMV). The internal ribosome entry site (IRES) enables expression of two unrelated reading frames from a single transcription unit. : packaging signal.

Full figure and legend (36K)

We transferred COL7A1 into cultured DEB keratinocytes and fibroblasts using the retroviral system. Transfection experiments showed that retroviral methods using pLI-COL and pDON-COL failed to introduce COL7A1 to DEB cells (data not shown). Plasmid pDON()-COL with the VSV-G system allowed a greater gene transfer after further concentration of the virus particles increased the transfer rate (Figure 2). Immunostaining revealed that the transfer rates in DEB keratinocytes and fibroblasts were almost the same (Figures 2 and 3a), and immunoblotting demonstrated that the amount of collagen VII in their culture media was also identical (Figure 3b). The average copy number per cell of the COL7A1 cDNA was evaluated by Southern blot analysis of genomic DNA extracted from the transduced cells. The result indicated that the intensities of 7.2 kb band from the integrated cDNA were the same in treated keratinocytes and fibroblasts, suggesting that integration copies for keratinocytes and fibroblasts was almost equal (Figure 3c). The copy number was estimated 2–3 by comparing with a serial dilution standard (data not shown).

Figure 2.

Successful gene transfer of COL7A1 into DEB fibroblasts and keratinocytes using retroviral systems. COL7A1 was transferred into DEB fibroblasts and keratinocytes using the retroviral systems. Immunostaining revealed that the transfer rates in DEB keratinocytes and fibroblasts were almost equal. The concentration of virus particles (by 10 or 50 times) using the VSV-G system improves the transfer rate. The values were represented the meanSD of six individual samples.

Full figure and legend (11K)

Figure 3.

Corrective gene transfer of the COL7A1 into DEB fibroblasts and keratinocytes. (a) Immnostaining showed that the gene-transfected DEB keratinocytes Kera (+) and fibroblasts Fib (+) expressed collagen VII, whereas no expression was found in either the untreated keratinocytes Kera (-) or fibroblasts Fib (-). Nuclei were counterstained with propidium iodide. (b) Western blot analysis demonstrated that the amount of transgene product in culture medium was almost the same between keratinocytes and fibroblasts. (c) Southern blot analysis of genomic DNA extracted from the transduced cells showed that the COL7A1 cDNA integration copies for keratinocytes and fibroblasts were almost equal.

Full figure and legend (100K)

In vivo COL7A1 expression in epidermis is higher than that in dermis

To determine the COL7A1 expression level in the epidermis and dermis in vivo, we separated the epidermis from the dermis, and measured COL7A1 mRNA levels using real-time PCR. Real-time PCR demonstrated that the COL7A1-specific signal (per RNA) of the epidermis was higher than that of the dermis by 3.2-fold (Figure 4). Comparison of the total RNA amounts from the epidermis and dermis in the same area of the excised normal skin showed that the amount from the dermis was higher than that from the epidermis by 3.4-fold (Figure 4). Thus, the quantity of COL7A1 expression in epidermis was significantly higher than that in dermis in vivo.

Figure 4.

In vivo COL7A1 expression in the epidermis is higher than that in the dermis. We measured COL7A1 mRNA levels in the epidermis (Epi) and dermis (Derm) in vivo. Real-time PCR demonstrated that the COL7A1-specific signal (per RNA) of the epidermis was higher than that of the dermis. Comparison of the total RNA amounts from the epidermis and dermis in the same area of the excised normal skin showed that the amount from the dermis was higher than that from the epidermis. The COL7A1 mRNA expression and total RNA amounts were expressed as an arbitrary scale. The values were represented the meanSD from three separate samples. ^*P<0.01, significant difference.

Full figure and legend (39K)

Gene-transferred fibroblasts can supply more collagen VII to the basement membrane zone than gene-transferred keratinocytes

We transplanted the gene-transferred DEB keratinocytes and fibroblasts into the wound of nude rats, and then observed COL7A1 deposition 3, 6, and 9 weeks after transplantation. In the skin graft with gene-transferred keratinocytes and untreated fibroblasts, the COL7A1 deposition was detectable in the basement membrane zone at 3 week and maintained this expression at least until 9 weeks (Figure 5). However, we found a greater accumulation of collagen VII in dermal–epidermal junction of the grafts using untreated keratinocytes and gene-transferred fibroblasts 3 weeks after transplantation. Furthermore, DEB fibroblasts transfected with COL7A1 demonstrated more dermal–epidermal junction collagen VII staining than COL7A1-transfected DEB keratinocytes/untreated fibroblast (Figure 5) from 6 to 9 weeks. The grafts of DEB keratinocytes and fibroblasts as controls demonstrated no deposition (Figure 5). Semiquantification of COL7A1 deposition in basement membrane zone in each point showed DEB fibroblasts with COL7A1 can supply higher amount of collagen VII to the basement membrane zone than DEB keratinocytes with COL7A1 (Figure 6). Significant differences were found between the keratinocytes and fibroblasts samples at 3W and 9W points.

Figure 5.

Gene-transferred fibroblasts can supply more collagen VII to the sub-basement membrane zone than gene-transferred keratinocytes. We transplanted gene-transferred DEB keratinocytes and DEB fibroblasts to nude rats wounded back skin, and observed COL7A1 deposition by immunohistochemistry at 3, 6, and 9 weeks after transplantation. The skin graft with gene-transferred keratinocytes and untreated fibroblasts (Kera 3W, Kera 6W, Kera 9W) started dermal–epidermal junction collagen VII deposition at 3 week and maintained it until 9 weeks. A greater accumulation of collagen VII in dermal–epidermal junction of the grafts using untreated keratinocytes and gene-transferred fibroblasts was found 3 weeks after transplantation (Fib 3W). The DEB fibroblasts transfected with COL7A1 demonstrated more dermal–epidermal junction collagen VII staining than COL7A1-transfected DEB keratinocytes/untreated fibroblast from 6 to 9 weeks (Fib 6W, Fib 9W). The control DEB keratinocyte and fibroblast cell grafts (Cont 3W, Cont 6W, Cont 9W) demonstrated no deposition. Arrowheads define the limit between the dermis and epidermis.

Full figure and legend (109K)

Figure 6.

Semiquantification of COL7A1 deposition in basement membrane zone. We transplanted gene-transferred DEB keratinocytes and DEB fibroblasts to nude rats wounded back skin, and observed COL7A1 deposition by immunohistochemistry at 3, 6, and 9 weeks after transplantation. To semiquantify COL7A1 deposition in basement membrane zone, we measured fluorescence intensity (arbitrary scale) in basement membrane zone at 10 areas at each point and the COL7A1 deposition index was expressed as the meanSD from the 10 values. K: the skin graft with gene-transferred keratinocytes and untreated fibroblasts; F: the graft with untreated keratinocytes and gene-transferred fibroblasts grafts; C: the graft with untreated keratinocyte and untreated fibroblasts. ^*P<0.01, significant difference.

Full figure and legend (12K)

Collagen VII released from gene-transferred cells forms ultrastructurally normal anchoring fibrils

We examined the ultrastructural formation of anchoring fibrils in the graft. The grafts with both gene-transferred keratinocytes and fibroblasts demonstrated filamentous loop structures just beneath the lamina densa, which were corresponding to anchoring fibrils (Figure 7). We could not see any filamentous structures in control (untransfected) samples without COL7A1.

Figure 7.

Collagen VII released from gene-transferred cells forms anchoring fibrils similar to normal skin. We examined the ultrastructural formation of anchoring fibrils in the grafts. The grafts with the gene-transferred keratinocytes (Kera) or the gene-transferred fibroblasts (Fib) demonstrated cross-banded, filamentous structures sometimes forming semicircular loops immediately beneath the lamina densa, corresponding to anchoring fibrils, whereas we failed to identify these filamentous structures in control without COL7A1 transfection (Cont).

Full figure and legend (50K)

Discussion

The developments in cloning the basement membrane protein genes have allowed the identification of the causative genes/proteins harboring the mutations responsible for this group of epidermolysis bullosa diseases (Uitto and Pulkkinen, 2001). We can now make good estimations about the prognosis and severity of these diseases with profound beneficial effects on genetic counseling and DNA-based prenatal diagnosis. However, patients most frequently desire an appropriate therapy for epidermolysis bullosa. Corrective transfer of the COL7A1 gene back into the skin cells is a promising treatment of DEB.

The previous Northern hybridization study revealed a high level of COL7A1 mRNA expression in cultured epidermal keratinocytes, whereas the expression was lower in cultured dermal fibroblasts (Ryynanen et al., 1992). These results indicate that epidermal keratinocytes and dermal fibroblasts express the collagen VII, but also suggest that epidermal keratinocytes are the primary source of collagen VII in developing human skin. Many investigators have utilized keratinocytes as the target cells of DEB gene therapy. Several methods including viral- (Ghazizadeh and Taichman, 2000) and non-viral-mediated transduction (Vogel, 2000) have been reported for in vivo and ex vivo gene transfer into keratinocytes. COL7A1 cDNA was recently transferred into cultured DEB keratinocytes using some methods including lentivirus- (Chen et al., 2002b), retrovirus- (Baldeschi et al., 2003) and C31 integrase-based approaches (Ortiz-Urda et al., 2002). The corrected DEB keratinocytes expressed the recombinant collagen VII and restored the in vivo synthesis of anchoring fibrils after implantation, demonstrating the feasibility of gene transfer using DEB keratinocytes. We also succeeded in transferring the COL7A1 into in vivo keratinocytes using the naked DNA method (Sawamura et al., 2002) although the COL7A1 transfer efficacy was lower than the above ex vivo method.

Many gene therapy protocols have already utilized retroviral vectors for clinical practices. In this study, we also showed a retroviral vector could transfer the 9 kb COL7A1 cDNA into DEB keratinocytes. Another group has succeeded in transducing COL7A1 gene to keratinocytes using a retroviral plasmid containing the Neo selection gene (Baldeschi et al., 2003). They showed that transduction efficacies to primary keratinocytes were 40 and 83–93% by retroviral vectors pLSRS-Ires-zero and pMSCV, respectively. However, the use of similar plasmids is not possible to efficiently introduce this gene into keratinocytes in our experiments. In our system, the efficacy was about 30%, which was lower than those in previous report. This study, as far as we know, has been the first to try a retronectin retroviral targeting system for keratinocytes. Retronectin is a recombinant peptide, which consists of three functional fibronectin domains and significantly enhances retrovirus-mediated gene transduction into mammalian cells. Our data showed that addition of retronectin increased transfer efficacy in keratinocytes by 3-fold (data not shown), indicating that retronectin is indeed efficient in this keratinocyte/retroviral system.

Some groups have succeeded in transferring this gene into keratinocytes as mentioned above. On the other hand, cutaneous injection of the DEB fibroblasts transduced using C31 integrase-based approach also restores collagen VII deposition along the dermal–epidermal junction (Ortiz-Urda et al., 2003). Also, gene-corrected DEB fibroblasts and normal human fibroblasts alone could supply type VII collagen deposition at the basement membrane zone in vivo (Woodley et al. 2003), and this implies a possibility that normal cultured human dermal fibroblasts are injected intradermally into recesive dystrophic epidermolysis bullosa patients' skin. Moreover, intradermal injection of lentiviral vector with COL7A1 increased collagen VII expression in fibroblasts and endothelial cells, resulting in stronger deposition of collagen VII along the basement membrane zone anchoring fibrils as seen by electron microscopy (Woodley et al., 2004). This study also introduced the COL7A1 gene into DEB fibroblasts using the retroviral method and the consequent collagen VII assembly beneath the basement membrane of the fibroblast containing graft.

In this study, we compared dermal fibroblasts and epidermal keratinocytes as efficient target recipient cells for the collagen VII transgene product. After retroviral introduction of COL7A1, the transfer efficacy and the amount of collagen VII in the cultured keratinocytes media supernatant are almost the same as those of fibroblasts. Interestingly, a series of skin graft experiments first demonstrated that gene-transfected fibroblasts more efficiently assembled collagen VII into the dermal–epidermal junction than the gene-transferred keratinocytes. Previous Northern blotting analysis revealed higher level of COL7A1 mRNA expression in cultured epidermal keratinocytes than fibroblasts (Ryynanen et al., 1992). This study utilized real-time PCR technique and confirmed that the epidermis produced much more collagen VII than the dermis in vivo. If gene-transferred fibroblasts and keratinocytes express similar amounts of type VII collagen also in vivo, the fibroblasts may have a better ability to supply type VII collagen to the basement membrane than the keratinocytes.

It is evident that expression of recombinant collagen VII is driven by heterologous promoters, which escape the regulatory mechanisms that govern the expression of endogenous collagen VII in the different cell types. Also, keratinocytes have been preferred because of the possibility they offer of targeting stem cells and compared to keratinocytes, fibroblasts rapidly senesce in vivo (Krueger, 2000). However, fibroblasts are more robust and less susceptible to growth arrest and differentiation than epidermal keratinocytes (Ortiz-Urda et al., 2003). Furthermore, genetically engineered fibroblasts have had their use explored for therapeutic applications including visceral and cutaneous implantation to supply gene products to circulation (Roth et al., 2001). Given the above combined factors, it is proposed that fibroblasts may be potentially more feasible and a better target of DEB gene therapy than keratinocytes.

Materials and Methods

Cell culture

Primary keratinocytes were isolated and grown in the presence of an irradiated 3T3 feeder layer (Rheinwald and Green, 1975). Briefly, keratinocytes, which were obtained from skin biopsy of a DEB patient and healthy controls, were cultured on feeder layers of mitomycin C-treated mouse 3T3 fibroblasts in DMEM: Ham's F-12 (3:1) supplemented with 10% fetal calf serum (FCS), 5 g/ml insulin, 10 ng/ml epidermal growth factor, 0.4 g/ml hydrocortisone, and 8 ng/ml cholera toxin. The DEB patient was diagnosed as the most severe subtype, and Hallopeau–Siemens type showed no COL7A1 expression in the skin and harbored heterozygous premature stop codon mutations 1474del8 and 5818delC. Fibroblasts were also obtained from skin biopsy from the DEB patient and healthy controls, and were cultured in DMEM with 10% FCS. Packaging cells amphopack-293 and GP2-293 (Clontech, Palo Alto, CA) were maintained in DMEM with 10% FCS, 2 mM glutamine, and 2 mM sodium pyruvate.

Informed consents were obtained from all individual subjects in this study. The protocols were approved by the Ethical Committee at Hokkaido University Graduate School of Medicine. This study was conducted according to the Declaration of Helsinki Principles.

Intrinsic expression of collagen VII in control keratinocytes and fibroblasts

Human skin was obtained from normal volunteers, and treated with 10 mg/ml dispase for 3 hours at 37°C to separate the epidermis from the dermis. The epidermal and dermal sheets were minced and total RNA was extracted using an RNeasy RNA extraction kit (Qiagen, Hilden, Germany). First-strand cDNA was synthesized with reverse transcriptase (Life Sciences Inc., St Petersburg, FL) using an oligo-dT primer. Assays-on-Demand™ Products for COL7A1 and GAPDH were purchased from Applied Biosystems (Foster City, CA). The 50 l reaction in each well contained 1 l of total cDNA, 300 nM sequence-specific primers, and 200 nM dual-labeled fluorogenic probe with 1 U of Taqman Universal PCR master mix (Applied Biosystems). A negative control PCR without template and a positive PCR control with a template of known amplification were included in each assay. The samples underwent the following stages: stage 1, 50°C for 2 minutes; stage 2, 95°C for 10 minutes; and stage 3, 95°C for 15 seconds, followed by 60°C for 1 minutes. Stage 3 was repeated 45 times. Gene-specific products were measured by means of an ABI Prism 7700 sequence detection system (Perkin-Elmer Applied Biosystems, Foster City, CA) continuously for 45 cycles. The COL7A1-specific signal was normalized by constitutively expressed GAPDH and expressed as arbitrary scale.

Construction of retroviral COL7A1 expression vectors and transfection

Human full-length COL7A1 cDNA was constructed from several overlapping cDNA clones (Sawamura et al., 2002). We employed two retroviral vectors, pLIXN (Clontech) and pDON-AI (Takara, Otsu, Japan), and full-length COL7A1 cDNA was inserted into the retroviral vectors to generate plasmids termed pLI-COL and pDON-COL, respectively (Figure 1). Also, we created pDON() by removing the SV-40 promoter and Neo gene from pDON-AI and constructed a retroviral vector with COL7A1 cDNA pDON()-COL (Figure 1). The recombinant retroviruses were produced by transfecting the retroviral plasmids into the amphotropic amphopack-293 packaging cells (Clontech) using calcium-phosphate co-precipitation. In addition, we tried VSV-G-pseudotyped retrovirus vectors. The retroviral plasmids and plasmid pVSV-G were cotransfected into pantropic GP2-293 packaging cells (Clontech). The viral particles were recovered from the cell culture medium 48 hours later and applied to keratinocyte or fibroblast cultures. To increase transfer efficacy, ultracentrifugation was performed to concentrate the VSV-G virus particles. The titer of the viral supernatant was determined by real-time quantitative PCR (Towers et al., 1999).

Cells infection with retrovirus

Keratinocytes and fibroblasts were cultured to up to 60% of confluency and then infected with the viral suspensions in 5 g/ml polybrene. To increase the virus–cell interactions, we coated the surface of the culture plates with 10 ng/ml retronectin (Takara; fibronectin fragment CH-296). After incubation for 24 hours at 32°C, we maintained the treated cells under fresh medium for another 24 hours until the transduction efficiency was assessed by immunofluorescence examination of the infected cells.

Immunostaining and immunoblot

Transfected cultured keratinocytes and fibroblasts were fixed with 2% parafromaldehye in phosphate-buffered saline, and were then incubated with the monoclonal antibody LH7.2 (1:100) against the NC1 domain of collagen VII (Chemicon, Temecula, CA) for 18 hours at 4°C. They were treated with secondary goat anti-mouse IgG antibodies conjugated with FITC (1:50) for 1 hour at 37°C, and preparations were examined under a fluorescence microscope. Nuclei were counterstained with propidium iodide (Dojindo Laboratories, Kumamoto, Japan). Subconfluent cell cultures were fed for 48 hours with serum-free medium supplemented with 50 g/ml ascorbic acid. For SDS-PAGE analysis, the culture medium was treated with Amicon Ultra-100,000 Centrifugal Filter Devices (Millipore, Bedford, MA) for concentration and desalting. The samples were separated on a 5% polyacrylamide gel under reducing conditions. Immunoblotting analysis was performed by treating with the LH7.2 monoclonal antibody (1:1,000) for 18 hours at 4°C, and then secondary goat anti-mouse IgG antibodies conjugated with peroxidase (1:2,000) for 1 hour at 37°C. The resultant complexes were processed for Phototope HRP Western Blot Detection System (Cell Signaling, Beverly, MA) according to the manufacturer's protocol.

Southern blot analysis

The average copy number per cell of the COL7A1 cDNA was evaluated by Southern blot analysis (Baldeschi et al., 2003). Briefly, genomic DNA was extracted from subconfluent cell cultures and digested with BglII and HindIII. Plasmid pDON()-COL was serially diluted with yeast genomic DNA at the final concentration ranging from 0.5 to 20 copies/cell. The digested DNA was electrophoresed on a 0.8% agarose gel and transferred to Zeta-Probe membrane (Bio-Rad, Hercules, CA). The 375 bp cDNA extending from exons 58 to 64 was generated by PCR amplification of COL7A1 cDNA as a template. This fragment was designed to recognize a 7.2 kb band from the integrated cDNA, and also 1.0 and 0.3 kb bands from the intrinsic COL7A1 gene. The membranes were hybridized with the 375 bp cDNA probe was labeled by random primed incorporation of digoxigenin-labeled 2'-deoxyuridine 5'-triphosphate using the DIG DNA Labeling Kit (Roche, Indianapolis, IN) according to the manufacturer's instructions. After high stringency washes, blots were visualized using an enhanced chemiluminescence system.

Grafting of gene-transferred DEB cells

Gene-transferred and untreated DEB keratinocytes and DEB fibroblasts were cultured using the above methods. Fibroblasts (10⁶) were seeded into a collagen sponge scaffold and maintained in DMEM with 10% FCS for 3 days. In nude rats (F344/N Jcl-rnu), the sites for transplantation were prepared by excising a 2 cm² area of dorsal epidermis and dermis, and then the collagen sponge (3 cm²) containing the fibroblasts was placed into the skin wound. The confluent cultures of 10⁶ keratinocytes were treated with dispase (1 nU/ml; Godo Shusei, Tokyo, Japan), and the floating epidermal sheet placed on the collagen sponge. Preliminary experiment showed that the number of fibroblasts was almost equal to that of keratinocytes when we applied the graft to the animal. An occlusive dressing was quickly placed over the graft to hold it in position and to prevent it from drying out and then the dressing was removed after 7 days. We prepared combinations of gene-transferred keratinocytes and untreated fibroblasts, of untreated keratinocytes and gene-transferred fibroblasts, and of untreated keratinocytes and fibroblasts as control. Skin biopsies were taken from the grafted skin at various time points and subjected to routine immunohistochemical staining using the LH7.2 monoclonal antibody and ultrastructural analysis. To semiquantify COL7A1 accumulation in basement membrane zone, we converted color images to gray-scale images, and measured fluorescence intensity (arbitrary scale) in basement membrane zone at 10 areas using NIH Image software. The COL7A1 deposition index was expressed as the meanSD from the 10 values.

Conflict of Interest

The authors state no conflict of interest.

References

1. Baldeschi C, Gache Y, Rattenholl A, Bouille P, Danos O, Ortonne JP et al. (2003) Genetic correction of canine dystrophic epidermolysis bullosa mediated by retroviral vectors. Hum Mol Genet 12:1897–1905 | Article | PubMed | ISI | ChemPort |
2. Burgeson RE (1993) Type VII collagen, anchoring fibrils, and epidermolysis bullosa. J Invest Dermatol 101:252–255 | Article | PubMed | ISI | ChemPort |
3. Chen M, Costa FK, Lindvay CR, Han YP, Woodley DT (2002a) The recombinant expression of full-length type VII collagen and characterization of molecular mechanisms underlying dystrophic epidermolysis bullosa. J Biol Chem 277:2118–2124 | Article | PubMed | ISI | ChemPort |
4. Chen M, Kasahara N, Keene DR, Chan L, Hoeffler WK, Finlay D et al. (2002b) Restoration of type VII collagen expression and function in dystrophic epidermolysis bullosa. Nat Genet 32:670–675 | Article | PubMed | ISI | ChemPort |
5. Christiano AM, Greenspan DS, Hoffman GG, Zhang X, Tamai Y, Lin AN et al. (1993) A missense mutation in type VII collagen in two affected siblings with recessive dystrophic epidermolysis bullosa. Nat Genet 4:62–66 | Article | PubMed | ISI | ChemPort |
6. Christiano AM, Greenspan DS, Lee S, Uitto J (1994a) Cloning of human type VII collagen, complete primary sequence of the alpha 1(VII) chain and identification of intragenic polymorphisms. J Biol Chem 269:20256–20262 | PubMed | ISI | ChemPort |
7. Christiano AM, Hoffman GG, Chung-Honet LC, Lee S, Cheng W, Uitto J et al. (1994b) Structural organization of the human type VII collagen gene (COL7A1), composed of more exons than any previously characterized gene. Genomics 21:169–179 | Article | PubMed | ISI | ChemPort |
8. Ghazizadeh S, Taichman LB (2000) Virus-mediated gene transfer for cutaneous gene therapy. Hum Gene Ther 11:2247–2251 | Article | PubMed | ISI | ChemPort |
9. Heinonen S, Mannikko M, Klement JF, Whitaker-Menezes D, Murphy GF, Uitto J (1999) Targeted inactivation of the type VII collagen gene (Col7a1) in mice results in severe blistering phenotype: a model for recessive dystrophic epidermolysis bullosa. J Cell Sci 112:3641–3648 | PubMed | ISI | ChemPort |
10. Krueger GG (2000) Fibroblasts and dermal gene therapy. Hum Gene Ther 11:2289–2296 | Article | PubMed | ISI | ChemPort |
11. Ortiz-Urda S, Lin Q, Green CL, Keene DR, Marinkovich MP, Khavari PA (2003) Injection of genetically engineered fibroblasts corrects regenerated human epidermolysis bullosa skin tissue. J Clin Invest 111:251–255 | Article | PubMed | ISI | ChemPort |
12. Ortiz-Urda S, Thyagarajan B, Keene DR, Lin Q, Fang M, Calos MP et al. (2002) Stable nonviral genetic correction of inherited human skin disease. Nat Med 8:1166–1170 | Article | PubMed | ISI | ChemPort |
13. Pulkkinen L, Uitto J (1999) Mutation analysis and molecular genetics of epidermolysis bullosa. Matrix Biol 18:29–42 | Article | PubMed | ISI | ChemPort |
14. Rheinwald JG, Green H (1975) Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6:331–343 | Article | PubMed | ISI | ChemPort |
15. Roth DA, Tawa NE Jr, O'Brien JM, Treco DA, Selden RF (2001) Factor VIII Transkaryotic Therapy Study Group: nonviral transfer of the gene encoding coagulation factor VIII in patients with severe hemophilia A. N Engl J Med 344:1735–1742 | Article | PubMed | ISI | ChemPort |
16. Ryynanen J, Sollberg S, Parente MG, Chung LC, Christiano AM, Uitto J (1992) Type VII collagen gene expression by cultured human cells and in fetal skin, abundant mRNA and protein levels in epidermal keratinocytes. J Clin Invest 89:163–168 | PubMed | ISI | ChemPort |
17. Sawamura D, Yasukawa K, Kodama K, Yokota K, Sato-Matsumura KC, Toshihiro T et al. (2002) The majority of keratinocytes incorporate intradermally injected plasmid DNA regardless of size but only a small proportion of cells can express the gene product. J Invest Dermatol 118:967–971 | Article | PubMed | ISI | ChemPort |
18. Towers GJ, Stockholm D, Labrousse-Najburg V, Carlier F, Danos O, Pages JC (1999) One step screening of retroviral producer clones by real time quantitative PCR. J Gene Med 1:352–359 | Article | PubMed | ISI | ChemPort |
19. Uitto J, Chung-Honet LC, Christiano AM (1992) Molecular biology and pathology of type VII collagen. Exp Dermatol 1:2–11 | PubMed | ChemPort |
20. Uitto J, Pulkkinen L (2001) Molecular genetics of heritable blistering disorders. Arch Dermatol 137:1458–1461 | PubMed | ISI | ChemPort |
21. Vogel JC (2000) Nonviral skin gene therapy. Hum Gene Ther 11:2253–2259 | Article | PubMed | ISI | ChemPort |
22. Woodley DT, Keene DR, Atha T, Huang Y, Ram R, Kasahara N et al. (2004) Intradermal injection of lentiviral vectors corrects regenerated human dystrophic epidermolysis bullosa skin tissue in vivo. Mol Ther 10:318–326 | Article | PubMed | ISI | ChemPort |
23. Woodley DT, Krueger GG, Jorgensen CM, Fairley JA, Atha T, Huang Y et al. (2003) Normal and gene-corrected dystrophic epidermolysis bullosa fibroblasts alone can produce type VII collagen at the basement membrane zone. J Invest Dermatol 121:1523–1747

Acknowledgments

We wish to thank Dr James R McMillan for proof reading and comments concerning this manuscript. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science, Sports, and Culture of Japan to D Sawamura (15390337, 17659331) and H Shimizu (15390336, 17209038), and by grants from the Ministry of Health of Japan to H Shimizu (H13-Specific Disease-02, H16-Intractable Disease-05).