Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 May 21.
Published in final edited form as: J Invest Dermatol. 2007 Dec 13;128(6):1476–1486. doi: 10.1038/sj.jid.5701197

PROTEIN THERAPEUTICS FOR JUNCTIONAL EPIDERMOLYSIS BULLOSA: INCORPORATION OF RECOMBINANT β3 CHAIN INTO LAMININ 332 IN β3-/- KERATINOCYTES IN VITRO

Olga Igoucheva 1, Aislinn Kelly 1, Jouni Uitto 1, Vitali Alexeev 1,*
PMCID: PMC3357058  NIHMSID: NIHMS376596  PMID: 18079746

Abstract

Junctional epidermolysis bullosa (JEB) is an inherited mechanobullous disease characterized by reduced adherence of the epidermal keratinocytes to the underlying dermis and often caused by the absence of functional laminin 332 due to the lack or dysfunction of its β3 chain. As there are no specific therapies for the JEB, we tested whether protein replacement strategy could be applicable for the restoration of the laminin 332 assembly and reversion of the JEB phenotype in human keratinocytes that lack β3 subunit. Here, we developed the protocol for production and purification of the biologically active recombinant β3 chain. Next, we demonstrated that delivery of recombinant β3 polypeptide into the endoplasmic reticulum of the immortalized β3-null keratinocytes led to the restoration of the laminin 332 assembly, secretion and deposition into the basement membrane zone, as confirmed by Western blot analysis, confocal immunofluorescent microscopy in vitro, and on cultured organotypic human JEB skin reconstructs. Although, the amount of laminin 332 produced by protein-treated β3-null keratinocytes is lower than that in normal human keratinocytes, our results demonstrate the applicability of the recombinant proteins for JEB treatment and open new perspectives for the development of novel therapeutics for this inherited, currently intractable, skin disorder.

Introduction

Inherited epidermolysis bullosa (EB) is a group of mechanobullous disorders, characterized by the development of the blisters following minor trauma to the skin (Fine et al., 2000). Traditionally EB is divided into three major categories: simplex, junctional and dystrophic, with at least twenty distinct clinical phenotypes. All forms of inherited EB result from mutations in the genes that encode structural proteins, which normally reside within the epidermis, dermal-epidermal junction, or the upper papillary dermis (Fine et al., 2000).

In the most severe junctional form of EB (JEB), the Herlitz variant (H-JEB), skin separation occurs at the level of lamina lucida of the cutaneous basement membrane zone (BMZ). This form of EB is frequently lethal during early postnatal period. Immunohistochemical analysis of the skin of the affected infants has demonstrated reduced or absent staining for laminin 3321 suggesting an important role for this protein in adhesion of basal keratinocytes to the basal lamina. Laminin 332 is composed of three distinct polypeptides, α3, β3 and γ2, which form a heterotrimer with a distinctive cross-shaped structure (Engel, 1992). Recent data suggest that assembly of the trimeric laminin proceeds from the formation of the β3γ2 heterodimer to the α3β3γ2 heterotrimer (Matsui et al., 1995a; Matsui et al., 1998) within rough endoplasmic reticulum (rER). Subsequently, trimeric laminin 332 is secreted from the keratinocytes and interacts with hemidesmosomal α6β4 integrin and the amino-terminal NC-1 domain of the type VII collagen, thus providing a firm attachment of keratinocytes to dermal anchoring fibrils (Chen et al., 1997, Brittingham et al., 2006). Therefore, the ablation of any subunit of laminin 332 results in separation of epidermis from dermis, as found in H-JEB.

Three distinctive genes, LAMA3, LAMB3, and LAMC2 encode the α3, β3, and γ2 chains of laminin 332, respectively. Mutational analysis of these genes has revealed that in most cases H-JEB is caused by premature stop codons in both alleles of any of the three laminin 332 genes. In less severe, type XVII collagen –independent, non-Herlitz variants of JEB with reduced laminin 332 expression, mutations in genes encoding laminin 332 polypeptides have been characterized mostly as misssense or in-frame splice mutations (Varki et al., 2006). These mutations do not cause the entire absence of individual polypeptides and apparently do not interfere with the assembly of the trimeric molecule, but rather disturb the function of the laminin 332, resulting in milder, non-lethal JEB phenotypes.

At present there are no specific therapies for any form of hereditary JEB. Advances in gene therapy hold much promise for the development of the gene-based molecular therapies of this inherited disorder (see Mavilio et al., 2006, Featherstone and Uitto 2007). However, current gene therapy approaches are not readily applicable for the treatment of patients with JEB, and the day-to-day management of the disease continues to revolve around prevention of mechanical trauma and infection. Therefore, the search for the new alternative approaches is warranted.

In recent years, recombinant proteins have emerged as a new class of molecular therapeutics. Recombinant proteins and polypeptides have been successfully used in treatment of a number of human diseases (Krejsa et al., 2006). Recently, Woodley and co-workers demonstrated the potential of protein application for treatment of cutaneous disorders, specifically the dystrophic forms of EB where skin separation below the lamina densa is caused by mutations in the COL7A1 gene (Woodley et al., 2004, Varki et al., 2007). In their study, human recombinant type VII collagen, injected intradermally into immunodeficient or immunocompetent mice, was retained in the BMZ for more than 2 months after a single injection of the protein. With respect to laminin 332, no studies using recombinant protein have been reported yet. However, Robbins and co-workers (Robbins et al., 2001a) using gene profiling approach, showed that culturing JEB keratinocytes on laminin 332-rich extracellular matrix restores the expression of many of the genes down-regulated in JEB.

In the current study, we extend the applicability of the protein therapy approach to a new group of blistering diseases, the junctional forms of EB and demonstrate that treatment of the H-JEB keratinocytes with recombinant β3 chain restores the formation and functional activity of the trimeric laminin 332.

Results

Expression vectors and protein production

To verify the expression of the full-length β3 subunit from pcDNA-LAMB3 expression vector, 2 μg of the plasmid were transfected into H-JEB keratinocytes using FuGene6 transfection reagent. The expression and secretion of the full-length β3 chain (140-kDa polypeptide) was confirmed by Western blot analysis using anti-β3-specific antibodies (Fig. 1a).

Fig. 1. Laminin β3 expression and secretion.

Fig. 1

Open in a new tab

Panel a – Expression and secretion of the laminin β3 from the pcDNA3.1-LAMB3 after transient transfection into H-JEB keratinocytes. Arrows on the left point to positions of molecular weight markers, arrows in the middle point to the β3 subunit. Panels b and c – GFP expression in 293F after transient transfection using 293Fectin transfection reagent (b) or Amaxa-based nucleofection (c). Panels d and e – Western blot analysis of the recombinant β3 chain secretion into culture media after transient transfection (d) or nucleofection (e) at various time points. Lane 1 – 48 hr, mock transfected 293F cells. Lanes 2, 3, and 4 - 24, 48, and 72 hr after pcDNA3.1-LAMB3 transfection/nucleofection, respectively. For the analysis of proteins by Western blotting, 5 μg of total protein from cell lysates and 15 μl of culture media were loaded per lane.

For the in vitro production of the recombinant protein, FreeStyle 293 Mammalian Expression System was used. This system allows large-scale protein production and mammalian post-translational modifications of the expressed proteins, important for the therapeutic applications. To investigate whether this expression system is suitable for the production of the recombinant β3 chain of laminin 332, the efficacy of plasmid DNA delivery into 293F cells was compared between liposome-based 293Fectin transfection reagent, provided as a part of the FreeStyle protein expression system, and electroporation-based Amaxa nucleofection. Side-by-side comparison clearly demonstrated that nucleofection of plasmid DNA resulted in at least 10-times higher DNA delivery than 293Fectin-based transfection (Fig. 1b and c). To determine which method of DNA delivery yields higher expression and secretion of the recombinant β3 polypeptide, culture media from pcDNA-LAMB3-transfected/nucleofected 293F cells (1×106 cells per reaction) were collected at different time points (24, 48 and 72 hr after transfection/nucleofection), concentrated, and analyzed by Western blotting for the presence of the secreted β3 subunit. As shown on figure 1, nucleofection resulted in a higher secretion of the recombinant protein into the culture medium during first 24 hrs, when compared with 293Fectin-mediated transfection (Fig. 1d and e). Remarkably, truncated and degraded proteins of the lower molecular weight were almost undetectable in the culture media collected from the nucleofected cells, and secretion of the β3 polypeptide remained high during additional 2 days of culture. Based on these findings, nucleofection was adopted as a primary DNA delivery method for the medium-scale production of the recombinant protein.

Medium-scale protein purification

To generate sufficient amount of the recombinant protein, 80 concurrent nucleofection reactions were performed. 293F cells nucleofected with the pcDNA-LAMB3 plasmid were cultured to obtain 3.2 L of culture media, which was concentrated and analyzed for the protein content by SDS-PAGE. Since no contaminating proteins in the range of ± 50 kDa from the β3 polypeptide were detected (Fig. 2a), the recombinant β3 was further purified by one-step gel-filtration chromatography using FPLC system. After multiple rounds of gel filtration, recombinant β3-containing fractions were combined, and an aliquot was analyzed by SDS-PAGE and Western blotting. As shown in figure 2, no contaminating proteins or degradation of the β3 chain could be detected in this fraction. Based on the protein concentration measurements by colorimetric assay and by absorbance at 280 nm, we were able to obtain approximately 50 μg of pure protein from 3.2 L of culture media.

Fig. 2. Purification and transfection of recombinant proteins.

Fig. 2

Open in a new tab

Panel a, lane 1 – concentrated culture media collected from β3 producing 293F cells; lane 2 - fraction containing laminin β3 after gel filtration. Samples of culture media and purified protein were prepared under denaturing condition, separated on 4-12% gradient PAGE, and stained with GelCode Blue reagent. Arrows on the left point to positions of molecular weight markers, arrow on the right points to the β3 subunit of laminin 332. Panels b, c, d, e – delivery of the recombinant β-galactosidase into normal (b, d) and H-JEB (c, e) keratinocytes. Panels b and c – photographs of protein transfected wells stained for the β-galactosidase activity; Panels d and e – magnified fields of panels b and c, respectively.

Protein delivery in vitro

After having achieved a relatively high yield and purity of the recombinant β3 chain of laminin 332, the in vitro protein delivery to normal human (NHK) and immortalized H-JEB keratinocytes was optimized using β-galactosidase reporter protein and ProJect™ protein transfection reagent. Under the optimized conditions, transfection of 4 μg of the recombinant protein per 5×105 initial “target” cells resulted in a highest efficacy of protein delivery, approaching 85% in both H-JEB and NHK cells (Fig. 2b-e).

Transfection of the recombinant β3 chain and restoration of the laminin 332 assembly

To determine whether recombinant protein can contribute to the assembly of the mature laminin 332, purified β3 subunit was transfected under optimized conditions into H-JEB keratinocytes. In all experiments, untreated H-JEB cells and NHK were used as negative and positive controls, respectively. After transfection, re-plating and additional 18 hr of culture, protein samples from cell lysates and corresponding culture media were prepared in native and denaturing conditions for Western blot analysis. Under both conditions, recombinant β3 polypeptide and chimeric laminin 332 that comprised of endogenous α3 and γ2 and recombinant β3 chains were detected in cell lysates and culture media of the protein-treated cells using β3-specific and laminin 332 conformation-specific GB3 antibodies, respectively (Fig. 3a and b). Neither one of the proteins were detected in untreated H-JEB keratinocytes. Laminin 332 was assembled as 460 kDa protein and secreted to the culture media where it was detected as processed 440 kDa protein (Fig. 3b). The presence of individual α3, β3, and γ2 chains in the assembled chimeric protein was confirmed by the Western blot analysis using chain-specific antibodies (Fig. 3c). The extent of the laminin 332 formation and secretion in β3-treated H-JEB cells was lower than that in NHK, suggesting that even the excess of the recombinant protein cannot fully restore trimeric protein assembly.

Fig. 3. Western blot and immunofluorescent analyses of the laminin β3 intracellular delivery and formation of the chimeric laminin 332.

Fig. 3

Open in a new tab

Panel a - analysis of laminin β3 delivery and secretion. Lane 1 and 2 – cell lysates, and lanes 3 and 4 - culture media collected from H-JEB and β3-treated H-JEB keratinocytes, respectively. Protein samples were prepared under denaturing conditions and detected on blots by laminin β3-specific antibodies. Arrow on the right points to β3 subunit. Panel b - Western blot analysis of assembly and secretion of laminin 332. Lane 1 and 2 – lysates and culture media collected from normal human keratinocytes, respectively; Lane 3 and 4 - lysates of H-JEB and β3-treated H-JEB keratinocytes, respectively; Lane 5 and 6 - culture media from H-JEB and β3-treated H-JEB keratinocytes, respectively. Protein samples were prepared in native conditions, separated on 4% PAGE and detected by laminin 332-specific (GB3) antibodies. Arrows on the right point to unprocessed and processed trimeric laminin 332 (α3β3γ2). Panel d – Western blot analysis of individual chains of laminin 332 in the trimeric protein, immunoprecipitated with GB3 antibodies. Arrows on the left point to individual chains of laminin 332, detected with protein-specific antibodies and to IgG. In all panels, arrows on the left point to positions of the molecular weight markers. Panel d – immuno-fluorescent detection of β3 and laminin 332 (red) in normal (NHK), H-JEB, and β3-treated H-JEB keratinocytes. Cell types are indicated above the panels, detected proteins indicated on the right. Nuclei stained with DPAI (blue). Scale bar = 100 μm. For the analysis of proteins by Western blotting, 5 μg of total protein from cell lysates and 15 μl of culture media were loaded per lane. The analysis was performed 18 hr after transfection.

Indirect immunofluorescent and confocal microscopy analyses

To confirm our findings at the cellular level, we analyzed the presence of the ProJect™-delivered β3 chain in treated keratinocytes. As shown in figure 3, in protein-transfected cultures approximately 65% of the cells contained β3 subunit in the cytoplasm in the peri-nuclear area, presumably in association with rER (Fig. 3d). Also, using GB3 antibodies, we confirmed the formation and deposition of the trimeric laminin 332 onto the plastic substrata from protein-treated H-JEB keratinocytes. Consistent with our Western blot analysis, the amount of the deposited laminin 332 was significantly lower than that from normal keratinocytes (Fig. 3c). Collectively, these data suggest that recombinant β3 chains contribute to the formation of mature laminin 332.

The intracellular delivery of the recombinant β3 chain and the formation of laminin 332 in protein-treated H-JEB keratinocytes were additionally confirmed by confocal microscopy. To determine whether the transfected protein can be delivered into the rER, where trimeric laminin is assembled (Nishiyama et al., 2000), we performed an indirect immunofluorescent detection of the β3 chain, laminin 332 and rER in protein-treated cells. Many of the protein-ProJect complexes formed large aggregates that did not penetrate the membranes of the cellular organelles (Fig. 4a). Nevertheless, similar to endogenously produced β3 chain in normal keratinocytes (Fig. 4k), recombinant β3 was detected in the majority of protein-treated cells in co-localization with rER (Fig. 4b and c). However, contrary to normal keratinocytes where trimeric laminin 332 was robustly co-localized with rER (Fig. 4l, m), in β3-treated cells only a small amount of the trimeric protein was associated with rER (Fig. 4e, f, i). The majority of the chimeric laminin 332 was evenly distributed in the cytoplasm and polarized towards the base of the cells (Fig. 4g, h, i). Moreover, it was organized in string-like structures that were radially extended from the nucleus to the periphery of the cells (Fig. 4g). Such intracellular organization was not detected in normal keratinocytes (Fig. 4l, m). When detected on plastic substrata, trimeric laminin 332 was widely spread over the tissue culture plate both in normal and protein-treated cells (Fig. 4j, and m, respectively). Although, the extent of substrata-associated fluorescence in β3-treated cells was lower than that in normal keratinocytes, these observations suggest that even a few protein-treated cells can secrete sufficient amounts of laminin 332 to restore the attachment of the neighboring β3-null keratinocytes.

Fig.4. Confocal immuno-fluorescent microscopy.

Fig.4

Open in a new tab

Panels a - c - Co-immunofluorescent detection of rough endoplasmic reticulum (rER) (red) and β3 subunit (green) in β3-transfected H-JEB keratinocytes; Panels d - j - Co-immunofluorescent detection of rER (red) and laminin 332 in protein-treated cells. Panel a - aggregates of β3-ProJect complexes and co-localization of β3 and rER; Panel a* - Z-stack of the consecutive Z-slices of cells on panel a; Panel b - rER in untransfected H-JEB keratinocytes and co-localization of β3 subunit and rER; Panel b* - Z-stack of the consecutive Z-slices of cells on panel b; Panel c - co-localization of β3 and rER lumen and tubules; Panels d - h consecutive Z-slices of a H-JEB keratinocytes transfected with recombinant β3 subunit. Panel i - Z-stack of the consecutive Z-slices of cells on panels d-h. Panel j - Deposition of laminin 332 from β3 treated keratinocytes. Panel k - co-localization of β3 subunit and ER in NHK; Panel l and m - co-localization of laminin 332 and ER in NHK and deposition of the trimeric protein. On panels a-c, and k: green arrowheads - laminin β3, yellow arrowheads - co-localization of the β3 subunit and ER, red arrowheads - rER in untransfected H-JEB cells. On panels d-i: green arrowheads - laminin 332, yellow arrowheads - co-localization of laminin 332 and ER, red arrowheads - ER in untransfected H-JEB cells. On panels j, j*, l, and m: green arrowheads with asterisk - secreted and deposited on plastic substrata laminin 332. Scale bar - 20 μm.

Collectively, these assays demonstrated that in vitro produced and purified β3 chain can be efficiently delivered into the β3-null keratinocytes and interact with endogenous α3 and γ2 chains to assemble mature laminin 332, which can be secreted from the keratinocytes and deposited onto plastic substrata.

Substrata binding and adhesion

As it has been shown previously (Vailly et al., 1998), restoration of the laminin β3 expression in H-JEB keratinocytes by gene transfer re-establishes keratinocytes attachment to the plastic substrata in vitro. To test, whether protein treatment also restores the attachment, β3-treated H-JEB keratinocytes were subjected to the substrata binding assay. At a relative centrifugal force of 500 × g, only 14% of normal human keratinocytes were dislodged, while 62% of the H-JEB cells lost attachment under similar conditions (Fig. 5a). However, substrate binding strength of the H-JEB cells after treatment was substantially increased, as indicated by lower number (32%) of detached cells after centrifugation (Fig. 5a). These results clearly indicate that treatment of the H-JEB keratinocyte with the recombinant β3 subunit results in markedly increased adhesion of the protein-treated cells.

Fig. 5. Substrata binding and adhesion assays, and analysis of the BMZ-associated laminin 332.

Fig. 5

Open in a new tab

Panel a - Substrata binding assay. Each point is the average of triplicates from 3 independent experiments presented as the mean ± SD of detached cells. NHK - normal human keratinocytes, H-JEB - β3-null keratinocytes, H-JEB + β3 - protein treated keratinocytes. Panel b - Cell adhesion assay. Each point is the average of triplicates from 3 independent experiments presented as the mean ± SD of absorbance at OD A280. H-JEB cells were seeded onto wells coated with extracellular matrix (ECM) produced by normal human keratinocytes (NHK), β3-null keratinocytes (H-JEB) and protein treated H-JEB keratinocytes (H-JEB + β3). Cells were allowed to attach for 2 or 12 hrs, as indicated below the panel. Panel c – quantitive, fluorescence-based analysis of the BMZ-associated laminin 332 in skin organotypic cultures comprised of normal human keratinocytes (NHK), H-JEB keratinocytes (H-JEB), or β3-treated keratinocytes (H-JEB + β3). Values are mean ± SD of the BMZ-associated fluorescence relative to total fluorescence (100%) of the analyzed fields.

To determine whether the extracellular matrix (ECM) secreted by protein-treated H-JEB keratinocytes can support cell adhesion, we utilized a modified adhesion assay (Carter et al., 1990). As shown in figure 5, during the first two hrs after plating, only few of the H-JEB keratinocytes were attached to matrices secreted by untreated H-JEB cells. At the same time, up to 20 times more keratinocytes were attached to matrices secreted by protein-treated (H-JEB+β3) and wild type keratinocytes (NHK), respectively (Fig. 5b). After 12 hr of incubation, virtually all seeded H-JEB cells (100%) were attached to matrices secreted by NHK cells, 72 % of cells were attached to matrices secreted by protein-transfected cells, and only 9% were attached to matrices secreted by H-JEB keratinocytes (Fig. 5b). These findings correlate with the data obtained in substrata binding assay and directly point to functionality of the chimeric laminin 332 in restoring adhesive properties of H-JEB keratinocytes.

Restoration of the laminin 332 in human skin organotypic cultures

To determine whether chimeric laminin 332 comprised of endogenous α3, γ2 and recombinant β3 chains can be secreted from protein treated H-JEB cells and deposited into the BMZ, we establish organotypic reconstructs of human skin. For the reconstruction of the epidermal component of the skin, NHK, H-JEB, and β3-treated treated H-JEB keratinocytes were used. All these cells gave rise to a multi-layered epidermis, which structurally resembled human skin (data not shown). Immunofluorescent analysis showed that β3 chain was present intracellularly in basal keratinocytes of normal skin and completely absent in H-JEB reconstructs (Fig. 6a and b). Recombinant β3 chain was also detected in a few of the protein treated H-JEB keratinocytes (Fig. 6c). The mature trimeric laminin 332 was present extracellularly as an evenly distributed layer along the BMZ in reconstructs of normal skin (Fig. 6d) and absent in H-JEB skin reconstructs (Fig. 6e). However, when β3-treated H-JEB keratinocytes were used for the reconstitution of the epidermis, mature laminin 332 was formed, secreted and deposited into the BMZ (Fig. 6f). Although the staining for laminin 332 was not continuous, trimeric protein was widely distributed throughout the reconstructs. Using AutoQuant imaging software we determined that on 50 consecutive cross-sections of a reconstruct that comprised β3-treated H-JEB keratinocytes (approximately 0.35 × 1.2 mm area, 1/200 of total reconstruct area), the largest gap between patches of laminin 332 was approximately 0.26 × 0.32 mm. The intensity of the AlexaFluor488 fluorescence that labels laminin 332 at the BMZ accounted for 54 % of that measured on the cross-sections of a normal human skin reconstruct (Fig. 5b).

Fig. 6. Organotypic cultures of normal, H-JEB, and β3-treated H-JEB human skin.

Fig. 6

Open in a new tab

Panels a-l – representative sections of human skin organotypic cultures stained for laminin β3 (a-c), laminin 332 (d-f), laminin 332 and α6 integrin (g-i), or laminin 332 and β4 integrin (j-i). Epidermal components of the organotypic cultures were reconstituted using normal human keratinocytes (NHK) (panels a, d, g, j), H-JEB keratinocytes (H-JEB) (panels b, e, h, k), and β3-reated keratinocytes (H-JEB + β3) (panels c, f, I, l). Scale bar - 100 μm.

As functional laminin 332 is important for hemidesmosomal formation and stability (Baker et al., 1996), we examined the effect of protein-based treatment on the polarization of the α6β4 integrin in human skin reconstructs after 4 weeks of the in vitro culture. During this extended culture period, reconstructs comprised of normal and protein-treated H-JEB keratinocytes showed clear BMZ-associated staining for the trimeric laminin 332, which was co-localized with both α6 and β4 integrins at the BMZ. In reconstructs of H-JEB skin, both α6 and β4 integrins were mislocolized throughout the epidermis in the absence of the β3 subunit (Fig. 6g-l). Taken together, these observations demonstrated that deposition of the chimeric laminin 332 into the BMZ is stable at least for 4 weeks and leads to proper polarization of both α6 and β4 integrins to the BMZ in protein-treated skin reconstructs.

Discussion

Our studies demonstrate that the recombinant β3 subunit of the human laminin 332 can be produced in vitro and used for the restoration of the trimeric laminin 332 assembly in immortalized β3-null keratinocytes. Similar to the laminin 332 produced by wild-type keratinocytes, the chimeric laminin 332 presumably undergoes extracellular maturation and restores the adhesion of the keratinocytes. Our data further suggest that recombinant individual chain of laminin 332 can be utilized for the restoration of trimeric laminin and, potentially, for treatment of JEB and alleviation and prevention of the blister formation associated with this devastating skin disorder.

During last two decades, various protein-based therapeutics have been developed and implemented into clinical practice for treatment of metabolic disorders, autoimmune diseases, and cancer. These therapeutics include drugs that either supplement (cytokines, growth factors, enzymes) or block (antibodies, soluble receptors, enzyme inhibitors) the activities of the endogenous proteins. The majority of them are recombinant proteins that in vivo can present their activities as monomers, homo-dimers or homo-multimers. In contrast, structural proteins, including those involved in the formation of the BMZ, such as collagens and laminins, represent specific proteins that have to undergo several significant modifications to become functionally active. For instance, to form mature laminin 332, all three individual chains of the protein (α3, β3, and γ2) have to go through intracellular post-translational modifications and assembly intoα3/β3/γ2 hetero-trimer. Furthermore, the trimeric protein has to be secreted and undergo extracellular processing prior to its association with type VII collagen and α6β4 integrin (Chen et al., 1997, Brittingham et al., 2006). The complexity of these events makes the development of protein-based therapeutics for the replacement of structural proteins challenging. However, considering limitations of the current gene therapy approaches, protein replacement seems to be an alternative strategy that, if successful, may provide immediate, clinically significant benefits for the JEB patients.

Most of the protein therapeutics are produced in mammalian cells. Production of recombinant proteins follows a well-established scheme that includes: (1) transfer of the recombinant DNA coding for a protein of interest; (2) selection and expansion of protein expressing cells; and (3) establishment of individual cell lines with the highest protein production. Here, utilizing transient expression system, rather than stable cell lines for protein production, we demonstrated that recombinant β3 chain of laminin 332 can be produced in relatively large quantities during a short period of time. We also showed that one-step gel-filtration can be optimized for the purification of the recombinant β3 chain to minimize protein degradation and increase the yield of the pure recombinant protein. However, if protein replacement strategy will show clinical benefits in upcoming pre-clinical studies on animal models of JEB (Kuster et al., 1997; Arin and Roop, 2001; Meng et al., 2003), alternative approaches for the production of recombinant chains of laminin 332 can be established. Currently, we are testing the use of strong mammalian promoters in conjunction with Universal Chromatin Opening Elements (UCOE's) and sodium byturate (Williams et al., 2005, de Poorter et al., 2007) for the enhancement of protein production.

A major question regarding protein replacement strategy for the restoration of laminin 332 is how to achieve effective delivery of the missing chain into defective keratinocytes in a manner that will facilitate formation of the trimeric molecule. Here we presented evidence that ProJect-mediated transfection of the H-JEB human keratinocytes in vitro with recombinant β3 chain resulted in the delivery of the protein into the rER and subsequent formation of the chimeric laminin 332. Although, the precise mechanism of ProJect-mediated protein delivery into the rER is not known, we suggest that protein-ProJect complexes fuse with the membranes of this organelle, releasing the protein into the rER. Although we found that β3-ProJect™ complexes often formed relatively large aggregates that could not be delivered into the rER we demonstrated that ProJect™-mediated transfection of β3 subunit into H-JEB keratinocytes leads to its incorporation into the trimeric laminin 332. Even so, for the future pre-clinical and clinical studies other protein transfection reagents, such as Chariot (Active Motif, Carlsbad, CA), or cell-penetrating oligo-arginine carriers (Bendifallah et al., 2006) and arginine-grafted polyamidoamonine dendrimers (Kim et al., 2006) can be tested.

Using confocal immunofluorescent microscopy, we also demonstrated that only a small portion of the trimeric laminin 332 in protein-transfected cells was associated with rER. The majority of the chimeric protein was widely distributed in the cytoplasm where it was organized into string-like structures. These observations suggest that immediately after incorporation of the recombinant β3 chain, assembled laminin 332 leaves the rER, becomes secreted from the cells, and deposits onto a relatively large area of the culture plate. Wide distribution of the chimeric laminin 332 on plastic substrate allows us to suggest that even relatively few protein-treated cells can produce functional laminin 332 in amounts sufficient to restore the adherence of the neighboring β3-negative keratinocytes. The data obtained in substrata binding assay, showing an essential restoration of the binding capacity of the H-JEB keratinocytes after protein treatment, directly support this notion.

To experimentally test whether protein-treated cells are capable of secreting sufficient amounts of laminin 332 to restore dermal-epidermal junction, we established skin organotypic cultures. As expected, individual β3 chains were detected inside basal keratinocytes of normal human skin reconstructs and were absent in H-JEB organotypic cultures. More importantly, immunofluorescent analyses of both normal and protein-treated H-JEB skin reconstructs showed polarized deposition of trimeric laminin 332 in the BMZ and proper polarization of the hemidesmosomal α6β4 integrins in the reconstituted epidermis. Although, laminin 332 deposition was uneven in reconstructs made of protein-treated cells, those patches were evenly distributed along the dermal-epidermal junction leaving only small gaps where the protein was absent. Remarkably, the quantity of the trimeric laminin produced by β3-treated H-JEB keratinocytes was determined to be for more than 50% of that produced by normal human keratinocytes during the same period in parallel reconstructs.

Collectively, our observations suggest two potential strategies of the protein-based therapeutics for JEB treatment/blister prevention: (1) direct in vivo transfection of the recombinant protein into epidermal keratinocytes, or (2) ex vivo transfection of the recombinant protein into keratinocytes isolated from JEB patient's skin and subsequent engraftment of the protein-treated cells onto the affected areas.

In self-renewing tissues, such as cutaneous epithelia, clinically useful therapies require sustainable presence of the therapeutics. Protein replacement strategy presents certain concerns regarding duration of the therapeutic effect, which, in general, depends on the stability of the recombinant protein at the therapeutic site. Previous studies showed that in human skin reconstructs, retrovirus-mediated gene transfer of the β3-encoding cDNA into H-JEB keratinocytes resulted in the presence of the restored laminin 332 at the BMZ for an extended period of time (up to 12 weeks), while β3-transduced cells were lost from the epidermis within 3-4 weeks (Robbins et al., 2001b). It was suggested that the extended presence of laminin 332 at the BMZ of the reconstructs is due to progeny of the transduced cells. However, these data may also suggest that incorporation of the protein into the BMZ leads to its stabilization and protection from rapid degradation by proteases. This notion is also supported by the observations that after single administration, recombinant human type VII collagen was stably present at the BMZ of mouse skin during a 2-month follow-up period (Woodley et al., 2004). Considering our data demonstrating the presence of the laminin 332 at the BMZ of the protein-treated H-JEB skin reconstructs for 2 to 4 weeks, here we hypothesize that laminin 332 replacement strategy may have long-lasting therapeutic effect. However, only in vivo animal studies, which are currently on-going in our laboratory, will allow us to experimentally address the question of duration of protein replacement and optimize treatment protocols to achieve durable therapeutic effects.

With further optimization of protein production and delivery, protein replacement strategies can be developed toward clinical application for localized treatment of JEB and other currently intractable genodermatoses.

Materials and Methods

Cell lines and culture conditions

Human keratinocytes isolated from a newborn patient suffering from the lethal, Herlitz type JEB were kindly provided by Dr. G. Meneguzzi and have been described elsewhere (Dellambra et al., 1998). These keratinocytes lack functional laminin 332 due to the absence of the β3 chain of laminin 332 caused by a homozygous premature stop codon in the LAMB3 gene. Primary H-JEB keratinocytes were immortalized using retrovirus-mediated transfer of the HPV16 E6E7 encoding gene. Viruses were collected from PA317 LXSN 16E6E7 cells (ATCC # CRL-2203). Primary normal human keratinocytes and fibroblasts were isolated from the skin biopsy of healthy donors as described previously (Nichols et al., 1977; Nickoloff et al., 1988). All keratinocytes were cultured in Epilife culture media with supplements and CaCl2 (Cascade Biologics, Portland, OR) on gelatinized plates (BD Bioscience, Rockford, MA). Fibroblasts were cultured in DMEM/F12 media supplemented with 15% FBS (Invitrogen, Carlsbad, CA).

Expression constructs

Full-length cDNA encoding human laminin β3 chain was obtained from total RNA isolated from normal human keratinocytes by RT-PCR using GeneRacer core kit and LAMB3-specific primers: direct 5’-ATGAGACCATTCTTCCTCTT-3’ and reverse 5’-CTTGCAGGTGGCATAGTAGA-3’. cDNA was subsequently inserted into the pcDNA 3.1 expression vector (Invitrogen, Carlsbad, CA). The integrity of the coding sequence was verified by direct DNA sequencing using gene-specific primers. The resulting plasmid was designated as pcDNA-LAMB3.

Optimization of DNA delivery into the 293F cells

Recombinant β3 chain was produced in 293F suspension cells cultured in a FreeStyle Expression Medium™ (Invitrogen, Carlsbad, CA). Optimization of DNA transfection into 293F cells was carried out using 293Fectin™ transfection reagent (Invitrogen, Carlsbad, CA) and Amaxa nucleofection reaction (Amaxa Inc, Köln, Germany). Two and 4 μg of the pMaxGFP plasmid encoding green fluorescent protein (Amaxa Inc, Köln, Germany) were used for transfection and nucleofection, respectively. Efficacy of the DNA delivery was estimated from GFP expression by fluorescent microscopy using AutoQuant imaging software (AutoQuant Imaging Inc., Troy, NY).

Production and purification of the recombinant protein

For the medium-scale protein production, 80 nucleofection reactions using 4 μg of plasmid DNA and 1×106 293F cells per reaction were set according to Amaxa protocol. Nucleofected cells were transferred into 200 ml of culture media to obtain 4×105 cells per ml and cultured for 24 hr. The volume of the media was then doubled, and selectable agent (G418) was added to the media in concentration 0.5 mg/ml. Each day for additional 4 days the volume of the culture media was doubled to obtain 3.2 L containing 1×106 protein-producing 293F cells per ml. The culture medium was collected, clarified by centrifugation, concentrated 200 times on Ultra-filtration Cell using YC-100 membrane (Millipore, Bedford, MA), and dialyzed against gel filtration start buffer containing 0.05 M Na2HPO4 and 0.15 M NaCl overnight in the presence of protease inhibitors at +4°C. Protein samples (4 ml each, 4% of the gel filtration column volume) were additionally purified by filtration and fractionated on HiLoad 16/60 Superdex 200 prep grade gel filtration column using Fast Performance Liquid Chromatography (FPLC) system (GE Healthcare Bio-Sciences, Piscataway, NJ). Individual protein-containing fractions were automatically collected based on the absorption at 280 nm. Protein content of each individual fraction was analyzed on SDS-PAGE. After multiple rounds of gel filtration, all fractions containing β3 chains of laminin 332 were combined, concentrated and stored at +4°C in the presence of 0.02% NaN3. Recombinant protein was dialyzed against PBS before the use.

Protein transfection in vitro

Normal and H-JEB keratinocytes (4×105 per well) were plated on a 6-well plates and 48 hr later, when cells reached 70 % confluence, the cultures were used for the optimization of protein delivery. Various amounts of β-galactosidase reporter protein (1, 2, or 4 μg) were transfected into keratinocytes using ProJect™ protein transfection reagent (Pierce, Rockford IL). Intracellular presence of the reporter protein was analyzed 12 hr after transfection using standard β-galactosidase colorimetric activity assay. The efficacy of the recombinant protein delivery was determined by counting β-galactosidase positive and negative cells on 5 random high-power fields (magnification x20) per transfection using AutoQuant imaging software.

For the detection of the recombinant β3 delivery and secretion of the restored laminin 332, H-JEB keratinocytes were transfected with 4 μg of the recombinant protein under optimal conditions. Six hr after transfection, protein-treated cells were harvested and re-plated onto gelatinized tissue culture plates. After additional 18 hr, tissue culture media from the control and protein-transfected cells were collected, cell debris was removed by centrifugation, and clarified media was concentrated 100 times on YM-100 CentriPrep concentrators (Millipore, Bedford, MA). Control and β3-transfected keratinocytes were lysed in a buffer containing 1% sodium deoxycholate, 1% Triton-X100, 0.1% SDS, 150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 1.5 mM EDTA, protease inhibitors and 1 mM PMSF. Protein concentration in media and cell lysates was measured by Coomassie blue colorimetric assay (Pierce, Rockford, IL). Samples for Western blot analysis were prepared under native and denaturing conditions using corresponding sample buffers (Pierce, Rockford, IL).

Western blot analysis

Native and denatured protein samples were separated on 4% (native conditions) and 4-12% gradient (denaturing conditions) PAGE gels at +4°C and at room temperature, respectively. Proteins were transferred onto PVDF membranes. β3 polypeptide was detected using chain-specific monoclonal antibodies raised against the C-terminal amino acids (1-159) of the protein (anti-Kalinin B1, BD Transduction Laboratories) in dilution 1:10,000. Mature trimeric laminin 332 was detected using GB3 monoclonal antibodies (Matsui et al., 1995a) in dilution 1:5,000. Immunocomplexes were detected by HRP-labeled secondary antibodies and visualized by WestFemto Supersignal detection kit (Pierce, Rockford, IL).

Indirect immuno-fluorescent analysis

For indirect immunofluorescent microscopy, control and protein-treated H-JEB keratinocytes were seeded onto gelatin coated chamber slides (Nalge Nunc, Rochester NY) and cultured as indicated in “Results” section. Prior to immunostaining, cells were fixed with methanol at –20°C and blocked with 1% BSA in Tris-buffered saline (TBS) for 1 hr at room temperature. Slides were incubated with β3- or laminin 332-specific antibodies in a dilution 1:200 overnight at +4°C. For the co-detection of the endoplasmic reticulum, ERAB antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) in dilution 1:200 were used. Immunocomplexes were detected with AlexaFluor488- or AlexaFluor594-conjugated secondary antibodies (Invitrogen, Carlsbad, CA). Slides were then examined under light fluorescent microscope Nikon TS100F or subjected to confocal fluorescent microscopy using Zeiss LSM 510 META Confocal, and confocal images were analyzed on the Zeiss LSM5 Image Browser software (Carl Zeiss GmbH, Germany). For the immunofluorescent detection of laminin β3 and laminin 332 on cross-sections of human skin organotypic cultures, 7 μm frozen sections were fixed and processed as described above for cultured cells.

Substrate binding assay

The binding strength of H-JEB cells to the plastic tissue culture substrata after protein treatment was quantified using modified McClay assay (McClay et al., 1981). Briefly, 6 hr after treatment with recombinant protein, H-JEB keratinocytes were removed from the plates by trypsinisation and re-plated onto 96-well plates in keratinocyte growth media at a density of 5×104 cells per well and kept for 24 hr. Plates were centrifuge at 500 × g at +4°C for 20 min. Cells adherent to the plastic substrata were fixed in 3% formaldehyde and stained with 0.1 % Crystal violet. After washes, plates were scanned at 560 nm on Bio Tech FL600 plate reader (Biotech Inc). Results were obtained from 3 independent experiments, 3 wells for each experiment.

Adhesion assay

To obtain ECM secreted by NHK, H-JEB, and protein-treated H-JEB, cells were seeded at density of 3×105 cells per well into uncoated 96-well plate. Cells were cultured for 3 days under standard tissue culture conditions. The adherent cells were then removed by incubation in 0.05% trypsin-EDTA for 5 min, 2 washes with trypsine inhibitor, incubation in 1% Triton X-100 in PBS for 5 min followed by overnight incubation at +4°C in PBS containing 20 mM EDTA and protease inhibitors. After incubation, detached cells were removed from the wells by pipetting. The ECM was digested with 10 μg/ml DNase I for 30 min in 1% PBS. The wells were washed with PBS and used for the assay. Untreated H-JEB keratinocytes were then plated onto ECM covered wells (5×105 cells per well) and incubated for 2 or 12 hr. Suspension cells were then removed by washing with PBS. Adherent cells were fixed and stained with Crystal violet, and the absorbance was measured at 560 nm as described above. Nonspecific background staining was evaluated on ECM-coated plates. The value of background staining for each ECM was subtracted from the value of staining in experimental wells. Results were obtained from three independent experiments, 3 wells for each experiment.

Organotypic human skin cultures

For dermal reconstruction, 1 ml of a cell-free buffered collagen solution consisting of bovine collagen type I (Organogenesis, Canton, MA) at a final concentration of 0.8 mg/ml in DMEM containing 10% fetal calf serum (FCS), Minimal Essential Medium with Earle's Salts, 200 mM L-glutamine, and 7.5% sodium bicarbonate was added to tissue reconstruct inserts (Organogenesis, Canton, MA). After hardening, the pre-coated acellular layer was overlaid with 3 ml of fibroblast-containing collagen solution (2.4 × 104 cells per ml). After 7 days of culturing, normal, H-JEB, or protein-transfected H-JEB keratinocytes (5×105 cells per reconstruct) were seeded on top of the contracted collagen gels. Reconstructs were cultured in complete keratinocyte culture medium (Cascade Biologics) with the addition of 1 ng/ml of human recombinant epidermal growth factor (EGF) for 2 days. The media were then replaced with the one containing reduced amount of EGF (0.2 ng/ml) and cells were cultured for additional 2 days. After that, medium was replaced with fresh one containing increased CaCl2 concentration (0.5 mM) without EGF. At this point, skin reconstructs were lifted to the air-liquid interface, and, unless otherwise stated, were maintained for additional 14 days with feeding from the bottom of the insert. Reconstructs were then harvested and embedded into OCT compound for cryo-sectioning and analysis.

Fluorescence imaging and analysis

For the analysis of laminin 332 distribution and quantitation, 50 consecutive cross-sections of the reconstructs comprised of normal, H-JEB, or β3-treated H-JEB keratinocytes were stained with GB3 antibodies and AlexaFluor488- labeled secondary antibodies. Images of stained cross-sections were taken and analyzed using AutoQuant imaging software (AutoQuant Imaging Inc., Troy, NY). For the analysis of distribution, all fluorescence-positive regions associated with the BMZ were automatically detected on each image. Consecutive images were merged into 3-dimensional composite. Fluorescence-positive regions associated with the BMZ were detected, and intensity of fluorescence was measured on each image using AutoQuant imaging software. Gaps between fluorescence-positive regions of the BMZ were manually outlined, and areas of the gaps were calculated using imaging software. Average value of the BMZ-associated fluorescence collected from 50 consecutive images was calculated. The average value of total fluorescence on 50 images of normal skin reconstruct was designated as 100%, and percentage of the BMZ-associated fluorescence in normal, H-JEB or β3-treted H-JEB skin reconstructs was calculated. Proteins on cryosections were detected using anti-laminin 332, anti-β3, anti-integrin α6 (Millipore, Bedford, MA), and anti-integrin β4 (Santa Cruz Biotechnology, Ink, Santa Cruz, CA) antibodies.

Acknowledgements

The authors would like to thank Drs. A. Fertala and M. Herlyn for helpful suggestions and discussions. All confocal immunofluorescent microscopy was done in the bioimaging facility of the Kimmel Cancer Center at Thomas Jefferson University. This research was supported by the NIH grant PO1AR038923 to J.U. and V.A., and by the NIH grant R21AR052508 to V.A.

Abbreviations used

JEB

Junctional Epidermolysis Bullosa

H-JEB

Herlitz variant of JEB

BMZ

basement membrane zone

rER

rough endoplasmic reticulum

ECM

extracellular matrix

Footnotes

1

Laminin 332 was previously known as laminin 5. For recently modified nomenclature of laminins see Aumailley et al., 2005

Literature sited

  1. Arin MJ, Roop DR. Disease model: heritable skin blistering. Trends Mol Med. 2001;7:422–424. doi: 10.1016/s1471-4914(01)02095-0. [DOI] [PubMed] [Google Scholar]
  2. Aumailley M, Bruckner-Tuderman L, Carter WG, Deutzmann R, Edgar D, Ekblom P, et al. A simplified laminin nomenclature. Matrix Biol. 2005;24:326–332. doi: 10.1016/j.matbio.2005.05.006. [DOI] [PubMed] [Google Scholar]
  3. Baker SE, Hopkinson SB, Fitchmun M, Andreason GL, Frasier F, Plopper G, et al. Laminin-5 and hemidesmosomes: role of the alpha 3 chain subunit in hemidesmosome stability and assembly. J Cell Sci. 1996;109(Pt 10):2509–2520. doi: 10.1242/jcs.109.10.2509. [DOI] [PubMed] [Google Scholar]
  4. Bendifallah N, Rasmussen FW, Zachar V, Ebbesen P, Nielsen PE, Koppelhus U. Evaluation of cell-penetrating peptides (CPPs) as vehicles for intracellular delivery of antisense peptide nucleic acid (PNA). Bioconjug Chem. 2006;17:750–758. doi: 10.1021/bc050283q. [DOI] [PubMed] [Google Scholar]
  5. Brittingham R, Uitto J, Fertala A. High-affinity binding of the NC1 domain of collagen VII to laminin 5 and collagen IV. Biochem Biophys Res Commun. 2006;343:692–699. doi: 10.1016/j.bbrc.2006.03.034. [DOI] [PubMed] [Google Scholar]
  6. Carter WG, Kaur P, Gil SG, Gahr PJ, Wayner EA. Distinct functions for integrins alpha 3 beta 1 in focal adhesions and alpha 6 beta 4/bullous pemphigoid antigen in a new stable anchoring contact (SAC) of keratinocytes: relation to hemidesmosomes. J Cell Biol. 1990;111:3141–3154. doi: 10.1083/jcb.111.6.3141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen M, Marinkovich MP, Veis A, Cai X, Rao CN, O'Toole EA, et al. Interactions of the amino-terminal noncollagenous (NC1) domain of type VII collagen with extracellular matrix components. A potential role in epidermal-dermal adherence in human skin. J Biol Chem. 1997;272:14516–14522. doi: 10.1074/jbc.272.23.14516. [DOI] [PubMed] [Google Scholar]
  8. Davern TJ. Molecular therapeutics of liver disease. Clin Liver Dis. 2001;5:381–414. vi. doi: 10.1016/s1089-3261(05)70171-9. [DOI] [PubMed] [Google Scholar]
  9. de Poorter JJ, Lipinski KS, Nelissen RG, Huizinga TW, Hoeben RC. Optimization of short-term transgene expression by sodium butyrate and ubiquitous chromatin opening elements (UCOEs). J Gene Med. 2007 doi: 10.1002/jgm.1057. in press. [DOI] [PubMed] [Google Scholar]
  10. Dellambra E, Vailly J, Pellegrini G, Bondanza S, Golisano O, Macchia C, et al. Corrective transduction of human epidermal stem cells in laminin-5-dependent junctional epidermolysis bullosa. Hum Gene Ther. 1998;9:1359–1370. doi: 10.1089/hum.1998.9.9-1359. [DOI] [PubMed] [Google Scholar]
  11. Engel J. Laminins and other strange proteins. Biochemistry. 1992;31:10643–10651. doi: 10.1021/bi00159a001. [DOI] [PubMed] [Google Scholar]
  12. Featherstone C, Uitto J. Ex vivo gene therapy cures a blistering skin disease. Trends Mol Med. 2007;13:219–222. doi: 10.1016/j.molmed.2007.03.006. [DOI] [PubMed] [Google Scholar]
  13. Fine JD, Eady RA, Bauer EA, Briggaman RA, Bruckner-Tuderman L, Christiano A, et al. Revised classification system for inherited epidermolysis bullosa: Report of the Second International Consensus Meeting on diagnosis and classification of epidermolysis bullosa. J Am Acad Dermatol. 2000;42:1051–1066. [PubMed] [Google Scholar]
  14. Kim JB, Choi JS, Nam K, Lee M, Park JS, Lee JK. Enhanced transfection of primary cortical cultures using arginine-grafted PAMAM dendrimer, PAMAM-Arg. J Control Release. 2006;114:110–117. doi: 10.1016/j.jconrel.2006.05.011. [DOI] [PubMed] [Google Scholar]
  15. Krejsa C, Rogge M, Sadee W. Protein therapeutics: new applications for pharmacogenetics. Nat Rev Drug Discov. 2006;5:507–521. doi: 10.1038/nrd2039. [DOI] [PubMed] [Google Scholar]
  16. Kuster JE, Guarnieri MH, Ault JG, Flaherty L, Swiatek PJ. IAP insertion in the murine LamB3 gene results in junctional epidermolysis bullosa. Mamm Genome. 1997;8:673–681. doi: 10.1007/s003359900535. [DOI] [PubMed] [Google Scholar]
  17. Matsui C, Nelson CF, Hernandez GT, Herron GS, Bauer EA, Hoeffler WK. Gamma 2 chain of laminin-5 is recognized by monoclonal antibody GB3. J Invest Dermatol. 1995a;105:648–652. doi: 10.1111/1523-1747.ep12324108. [DOI] [PubMed] [Google Scholar]
  18. Matsui C, Pereira P, Wang CK, Nelson CF, Kutzkey T, Lanigan C, et al. Extent of laminin-5 assembly and secretion effect junctional epidermolysis bullosa phenotype. J Exp Med. 1998;187:1273–1283. doi: 10.1084/jem.187.8.1273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Matsui C, Wang CK, Nelson CF, Bauer EA, Hoeffler WK. The assembly of laminin-5 subunits. J Biol Chem. 1995b;270:23496–23503. doi: 10.1074/jbc.270.40.23496. [DOI] [PubMed] [Google Scholar]
  20. Mavilio F, Pellegrini G, Ferrari S, Di Nunzio F, Di Iorio E, Recchia A, et al. Correction of junctional epidermolysis bullosa by transplantation of genetically modified epidermal stem cells. Nat Med. 2006;12:1397–1402. doi: 10.1038/nm1504. [DOI] [PubMed] [Google Scholar]
  21. McClay DR, Wessel GM, Marchase RB. Intercellular recognition: quantitation of initial binding events. Proc Natl Acad Sci U S A. 1981;78:4975–4979. doi: 10.1073/pnas.78.8.4975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Meng X, Klement JF, Leperi DA, Birk DE, Sasaki T, Timpl R, et al. Targeted inactivation of murine laminin gamma2-chain gene recapitulates human junctional epidermolysis bullosa. J Invest Dermatol. 2003;121:720–731. doi: 10.1046/j.1523-1747.2003.12515.x. [DOI] [PubMed] [Google Scholar]
  23. Nichols WW, Murphy DG, Cristofalo VJ, Toji LH, Greene AE, Dwight SA. Characterization of a new human diploid cell strain, IMR-90. Science. 1977;196:60–63. doi: 10.1126/science.841339. [DOI] [PubMed] [Google Scholar]
  24. Nickoloff BJ, Mitra RS, Riser BL, Dixit VM, Varani J. Modulation of keratinocyte motility. Correlation with production of extracellular matrix molecules in response to growth promoting and antiproliferative factors. Am J Pathol. 1988;132:543–551. [PMC free article] [PubMed] [Google Scholar]
  25. Nishiyama T, Amano S, Tsunenaga M, Kadoya K, Takeda A, Adachi E, et al. The importance of laminin 5 in the dermal-epidermal basement membrane. J Dermatol Sci. 2000;24(Suppl 1):S51–59. doi: 10.1016/s0923-1811(00)00142-0. [DOI] [PubMed] [Google Scholar]
  26. Robbins PB, Sheu SM, Goodnough JB, Khavari PA. Impact of laminin 5 beta3 gene versus protein replacement on gene expression patterns in junctional epidermolysis bullosa. Hum Gene Ther. 2001a;12:1443–1448. doi: 10.1089/104303401750298599. [DOI] [PubMed] [Google Scholar]
  27. 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 U S A. 2001b;98:5193–5198. doi: 10.1073/pnas.091484998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Vailly J, Gagnoux-Palacios L, Dell'Ambra E, Romero C, Pinola M, Zambruno G, et al. Corrective gene transfer of keratinocytes from patients with junctional epidermolysis bullosa restores assembly of hemidesmosomes in reconstructed epithelia. Gene Ther. 1998;5:1322–1332. doi: 10.1038/sj.gt.3300730. [DOI] [PubMed] [Google Scholar]
  29. Varki R, Sadowski S, Pfendner E, Uitto J. Epidermolysis bullosa. I. Molecular genetics of the junctional and hemidesmosomal variants. J Med Genet. 2006;43:641–652. doi: 10.1136/jmg.2005.039685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Varki R, Sadowski S, Uitto J, Pfendner E. Epidermolysis bullosa. II. Type VII collagen mutations and phenotype-genotype correlations in the dystrophic subtypes. J Med Genet. 2007;44:181–192. doi: 10.1136/jmg.2006.045302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Williams S, Mustoe T, Mulcahy T, Griffiths M, Simpson D, Antoniou M, et al. CpG-island fragments from the HNRPA2B1/CBX3 genomic locus reduce silencing and enhance transgene expression from the hCMV promoter/enhancer in mammalian cells. BMC Biotechnol. 2005;5:17. doi: 10.1186/1472-6750-5-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Woodley DT, Keene DR, Atha T, Huang Y, Lipman K, Li W, et al. Injection of recombinant human type VII collagen restores collagen function in dystrophic epidermolysis bullosa. Nat Med. 2004;10:693–695. doi: 10.1038/nm1063. [DOI] [PubMed] [Google Scholar]

RESOURCES