Characterization of the initial response of engineered human skin to sulfur mustard.

We have used a new approach to identify early events in sulfur mustard-induced, cutaneous injury by exposing human, bioengineered tissues that mimic human skin to this agent to determine the morphologic, apoptotic, inflammatory, ultrastructural, and basement membrane alterations that lead to dermal-epidermal separation. We found distinct prevesication and post-vesication phases of tissue damage that were identified 6 and 24 h after sulfur mustard (SM) exposure, respectively. Prevesication (6 h) injury was restricted to small groups of basal keratinocytes that underwent apoptotic cell death independent of SM dose. Immunoreactivity for basement membrane proteins was preserved and basement membrane ultrastructure was intact 6 h after exposure. Dermal-epidermal separation was seen by the presence of microvesicles 24 h after SM exposure. This change was accompanied by the dose-dependent induction of apoptosis, focal loss of basement membrane immunoreactivity, increase in acute inflammatory cell infiltration, and ultrastructural evidence of altered basement membrane integrity. These studies provide important proof of concept that bioengineered, human skin demonstrates many alterations previously found in animal models of cutaneous SM injury. These findings further our understanding of mechanisms of SM-induced damage and can help development of new countermeasures designed to limit the morbidity and mortality caused by this chemical agent.

Basement membrane proteins promote progression of intraepithelial neoplasia in 3-dimensional models of human stratified epithelium.

We have developed novel 3-dimensional in vitro and in vivo tissue models that mimic premalignant disease of human stratified epithelium in order to analyze the stromal contribution of extracellular matrix and basement membrane proteins to the progression of intraepithelial neoplasia. Three-dimensional, organotypic cultures were grown either on a de-epidermalized human dermis with pre-existing basement membrane components on its surface (AlloDerm), on a Type I collagen gel that lacked basement membrane proteins or on polycarbonate membranes coated with purified extracellular matrix proteins. When tumor cells (HaCaT-II4) were mixed with normal keratinocytes (4:1/normals:HaCaT-II4), tumor cells selectively attached, persisted and proliferated at the dermal-epidermal interface in vitro and generated dysplastic tissues when transplanted to nude mice only when grown in the presence of the AlloDerm substrate. This stromal interface was permissive for tumor cell attachment due to the rapid assembly of structured basement membrane. When tumor cells were mixed with normal keratinocytes and grown on polycarbonate membranes coated with individual extracellular matrix or basement membrane components, selective attachment and significant intraepithelial expansion occurred only on laminin 1 and Type IV collagen-coated membranes. This preferential adhesion of tumor cells restricted the synthesis of laminin 5 to basal cells where it was deposited in a polarized distribution. Western blot analysis revealed that tumor cell attachment was not due to differences in the synthesis or processing of laminin 5. Thus, intraepithelial progression towards premalignant disease is dependent on the selective adhesion of cells with malignant potential to basement membrane proteins that provide a permissive template for their persistence and expansion.