Laminin 5 is a basement membrane component that actively promotes adhesion and migration of epithelial cells. Laminin 5 undergoes extracellular proteolysis of the γ2 chain that removes the NH2-terminal short arm of the polypeptide and reduces the size of laminin 5 from 440 to 400 kD. The functional consequence of this event remains obscure, although lines of evidence indicate that cleavage of the γ2 chain potently stimulated scattering and migration of keratinocytes and cancer cells. To define the biological role of the γ2 chain short arm, we expressed mutated γ2 cDNAs into immortalized γ2-null keratinocytes. By immunofluorescence and immunohistochemical studies, cell detachment, and adhesion assays, we found that the γ2 short arm drives deposition of laminin 5 into the extracellular matrix (ECM) and sustains cell adhesion. Our results demonstrate that the unprocessed 440-kD form of laminin 5 is a biologically active adhesion ligand, and that the γ2 globular domain IV is involved in intermolecular interactions that mediate integration of laminin 5 in the ECM and cell attachment.

Recent achievements in the genetic correction of keratinocytes isolated from patients with junctional epidermolysis bullosa have paved the way to a gene therapy approach for the disease. Because gene therapy protocols require preclinical validation in animals, we have characterized spontaneous animal models of junctional epidermolysis bullosa. In this study we have elucidated the genetic basis of the hereditary junctional mechanobullous disease in the Belgian horse, a condition characterized by blistering of the skin and mouth epithelia, and exungulation (loss of the hoof). Immunofluorescence analysis associated the condition to the absent expression of the gamma2 chain of laminin 5 and designated Lamc2 as the candidate gene. Comparative analysis of the nucleotide sequence of the full-length gamma2 cDNA isolated by reverse transcription polymerase chain reaction amplification of total RNA purified from the epithelium of a junctional epidermolysis bullosa foal and a healthy control disclosed a homozygous basepair insertion (1368insC) in the affected animal. Mutation 1368insC results in a downstream premature termination codon and is predicted to cause absent expression of the laminin gamma2 polypeptide. Our results also show that: (i) the horse junctional epidermolysis bullosa genetically corresponds to the severe Herlitz form of junctional epidermolysis bullosa in man; (ii) the amino acid sequence and structure of the horse laminin gamma2 chain are virtually identical to the human counterpart; (iii) the moderate eruption of skin blisters in the affected animals with respect to the human Herlitz junctional epidermolysis bullosa patients correlates with the protection provided by hair. Our observations suggest that the affected foals are a convenient source of epithelial cells from tissues that cannot be obtained from human junctional epidermolysis bullosa patients, and imply that hairless strains of animals with recessive skin disorders would be the best models for in vivo gene therapy approaches to skin blistering diseases.

Junctional epidermolysis bullosa (JEB) is a genodermatosis suitable for gene therapy because conventional treatments are ineffective. Here, we elucidate the genetic basis of mild JEB in a breed of dogs that display all the clinical traits observed in JEB patients. The condition is associated with reduced expression of laminin 5 caused by a homozygous insertion (4818+207ins6.5 kb) of repetitive satellite DNA within intron 35 of the gene (lama3) for the laminin alpha3 chain. The intronic mutation interferes with maturation of the alpha3 pre-messenger RNA resulting in the coexpression of a transcript with a 227 nucleotide insertion and a wild-type mRNA that encodes scant amounts of the alpha3 polypeptide. Our results show that the amino acid sequence and structure of the canine and human alpha3 chain are highly conserved and that the reduced expression of laminin 5 affects the adhesion and clonogenic potential of the JEB keratinocytes. These JEB dogs provide the opportunity to perform gene delivery in a naturally occurring genodermatosis and to evaluate host tolerance to recombinant laminin 5.

The easy accessibility of the skin as a therapeutic target provides an exciting potential for this organ for the development of gene therapy protocols for cutaneous diseases and a variety of metabolic disorders. Thus far, full phenotypic reversion of a diseased phenotype has been achieved in vivo for junctional epidermolysis bullosa and X-linked or lamellar ichthyosis and in vitro for xeroderma pigmentosum. These recessive skin diseases are characterized by skin blistering, abnormalities in epidermal differentiation and increased development of skin cancers, respectively. Corrective gene delivery at both molecular and functional levels was achieved by transduction of cultured skin cells using retroviral vectors carrying the specific curative cDNA. These positive results should prompt clinical trials based on transplantation of artificial epithelia reconstructed ex vivo using genetically modified keratinocytes. Promising results have also been obtained in phenotypic reversion of cells isolated from patients suffering from a number of metabolic diseases such as gyrate atrophy, familial hypercholesterolemia or phenylketonuria. In these diseases transplantation of autologous artificial epithelia expressing the transgenes of interest or direct transfer of the DNA to the skin represents a potential therapeutic approach for the systemic delivery of active molecules. Successful cutaneous gene therapy trials, however, require development of protocols for efficient gene transfer to epidermal stem cells, and information about the host immune response to the recombinant polypeptides produced by the implanted keratinocytes. The availability of spontaneous animal models for genodermatoses will validate the gene therapy approach in preclinical trials.