9319560 Coulombe Keratin filament assembly begins with the formation of an heterodimer of type I and type II keratins through coiled-coil interactions along their 46 nm long alpha-helical domain. Two rod-shaped dimer molecules then interact laterally with an antiparallel orientation to form a tetramer, which is the only keratin assembly intermediate that can be isolated in high yields. Although epithelial cells either actively synthesize or contain multiple keratins, we do not yet understand the principles and mechanisms controlling the pairing of type I and type II keratins in vivo. The aim of this project is to characterize this, by focusing on two natural type I - type II keratin "pairs" which under certain conditions co-exist in epidermal cells: K5-K14 and K6-K16/K17. The keratin composition of populations of native filaments isolated from epidermal and tongue epithelial keratinocytes cultured under various conditions will be determined, using ultrastructural immunocytochemistry combined with a classical biochemical and immunochemical approach. A combination of chromatography, chemical cross-linking and immunochemistry will be used to determine the keratin composition of tetramer and dimer subunits derived from native keratin filaments. In addition, the composition of the heterotypic complexes that form when multiple type I and type II keratins are co-translated in a rabbit reticulocyte lysate will be determined. The combination of the two sets of data, i.e., the keratin composition of native filament subunits vs. that of filament subunits generated by co-translation of the same keratin mRNAs in a cell-free extract, will reveal whether pairing between keratins in vivo is dictated by the natural affinities between type I and type II keratins, or alternatively, is an assisted and thus regulated process. In addition, chimeric proteins (e.g. K14-K16 hybrids) will be generated, followed by site-specific mutagenesis, in order to address which sequence element(s) are responsible for th e pairing preferences between type I and type II keratins. These experiments will establish the type I - type II pairing preferences among two major keratin pairs and will provide insights into the sequence motifs and mechanisms controlling coiled-coil dimer formation in vivo. %%% Keratins are the epithelial-specific intermediate filament (IF) proteins. They assemble into cytoskeletal filaments of 10 nm diameter that are organized inside cells into a supramolecular network. The more than thirty keratin genes in humans have been subdivided into two distinct types (I and II) based on sequence homology, and their expression in epithelia is regulated in a tissue-type and differentiation-specific manner. Since keratin filaments are obligate heteropolymers constructed from type I and type II monomers in a 1:1 molar ration, epithelial cells express at least one member of each sequence subtype. Many of the keratins are coordinately expressed as natural type I -type II "pairs" that are associated with a well-defined epithelial context in mammals. Other keratins lack a "natural" partner and a predictable pattern of expression. In spite of these remarkable features, there is very little data pertaining to the functional relevance of keratin sequence diversity, and its impact on the structure and function of epithelial cells and tissues. The concept of "natural keratin pairs" stemmed from the notion that some type I and type II keratin genes are co-regulated at the transcriptional level. However, that does not imply exclusive co-polymerization of the corresponding proteins into 10 nm filaments in vivo. In fact, this possibility has never been thoroughly addressed, even though there are many natural instances in which an epithelial cell features a multiplicity of type I and type II keratins . The potential for such "mixing" at the 10 nm filament level is strengthened by recent studies indicating that keratin filaments are dynamic in vivo, with exchange between a soluble pool of subunits and the polymer (filament) pool. The mechanisms controlling the pairing of type I and II keratins as they are translated, as well as in a post-translational context that is dynamic, have yet to be investigated. The aim of this project is to fill this knowledge gap. In addition to the obvious importance of this kind of information for the understanding of how the intermediate filament networks of cells are assembled, this knowledge will be useful for exploiting such self-assembling nanoscale biological materials either directly or biomimetically for a variety of practical applications. ***
