Polarity and gradients in lepidopteran wing epidermis. II. The differential adhesiveness model: gradient of a non-diffusible cell surface parameter.

For explaining the Manduca wing gradient (Nardi & Kafatos, 1976) a model which postulates a proximo-distal gradient in cellular adhesiveness is considered. The model is based on Steinberg's (1963) differential adhesiveness hypothesis. Rosette formation in certain transposed and/or reoriented grafts can be adequately explained by this model. Several predictions, formulated by using the concept of surface free energy as a thermodynamic measure of adhesiveness, have been tested and proven correct. (1) Transposed grafts tend to assume circular forms, which are configurations of minimum free energy. (2) Because of the pressure difference expected across the interface of two cell populations with different surface free energies, cell densities increase in both distally and proximally transposed grafts. As a corollary to this rule, final size of a graft is a function of its distance from the original position. (3) Histological sections of host-graft boundaries suggest minimal cell contact at the interface. In proximal grafts placed in distal regions, cell density is far lower near the host-graft interface, as compared to the high interior density; the peripheries of distal grafts do not show this effect. (4) Juxtaposition of three different wing regions in all possible arrangements yields the expected two-dimensional configurations. (5) Differences in adhesiveness can be demonstrated by allowing two different wing grafts to interact in an essentially neutral environment (i.e. at a leg or antenna site). as the distance between two given graft regions increases, the extent of their final contact decreases. When applied to other insect systems, the model not only offers an alternative interpretation for results currently explained by diffusible substance models, but also accounts for certain features that were unexplained by other models.