A model for the genetic control of differential adhesion in insect epidermis

Abstract

Adhesive relations among cells are believed to play a major role in determining patterns of serial homology, of intercalary regeneration, and of neuronal connectivity. Models for the genetic control of adhesion during development can provide a framework for further analysis of these phenomena. Investigators studying development of Drosophila have proposed that differentiation of segments and of imaginal discs is controlled by a set of bistable “selector genes”. In each region the settings of the selector genes form a binary “word” which determines the properties of cells in the region, including their adhesiveness. I have made an explicit proposal for the relation between binary words and adhesiveness, by assuming that active selector genes repress synthesis of “adhesor” macromolecules, which promote adhesion. This hypothesis correctly predicts the relative cohesiveness of cells in four pupal tissues of the moth Manduca. Works of cohesion and adhesion among the four cell types are deduced from published results of grafting experiments by modelling insect epidermis as a viscoelastic fluid. Further comparisons between deductions from the genetic and fluid models suggest that selector genes, or the adhesor molecules they regulate, interact within single cells in determining adhesiveness between cells. From a specific version of the genetic model I deduce that pairwise interactions between selector genes or adhesor molecules can determine many, though not all, of the relative works of adhesion between unlike cells in Manduca. The genetic and fluid models thus provide a set of working hypotheses for predicting patterns of intercellular adhesion in insect epidermis and for analyzing results of experiments designed to test such predictions.