Development

The larval development of the Drosophila melanogaster wings is organized by the protein Wingless, which is secreted by cells adjacent to the dorsal–ventral (DV) boundary. Two signaling processes acting between the second and early third instars and between the mid- and late third instar control the expression of Wingless in these boundary cells.

A preliminary model (Apterous-dependent network) was presented in (see [1]), for the inter-cellular regulatory network activating Notch at the dorsal–ventral boundary in the wing imaginal disc of Drosophila. This model focussed on the cross-regulations between five genes (within and between two cells), which implements the dorsal–ventral border in the developing imaginal disc.

A refined model (complete network) was then published (see [2]) that integrates both signaling processes into a logical multivalued model encompassing four cells, i.e., a boundary and a flanking cell at each side of the boundary. Computer simulations of this model enable a qualitative reproduction of the main wild-type and mutant phenotypes described in the experimental literature. During the first signaling process, Notch becomes activated by the first signaling process in an Apterous-dependent manner. In silico perturbation experiments show that this early activation of Notch is unstable in the absence of Apterous. However, during the second signaling process, the Notch pattern becomes consolidated, and thus independent of Apterous, through activation of the paracrine positive feedback circuit of Wingless. Consequently, we propose that appropriate delays for Apterous inactivation and Wingless induction by Notch are crucial to maintain the wild-type expression at the dorsal–ventral boundary. Finally, another mutant simulation shows that cut expression might be shifted to late larval stages because of a potential interference with the early signaling process.

References

Initially activated by the pair-rule genes, the expression patterns of the segment polarity genes engrailed and wingless become consolidated through inter-cellular interactions between juxtaposed cells. Here, we delineate a logical model focusing on a dozen molecular components at the core of the regulatory network controlling this process. Our model analysis leads to the following conclusions: (1) the pair-rule signals, which activate engrailed and wingless genes independently of each other, need to be operative until the inter-cellular circuit involving these two genes is functional. This implies that the pair-rule pattern is instrumental both in determining the activation of the genes engrailed and wingless in rows of adjacent cells, and in consolidating these expression patterns. (2) The consolidation of engrailed and wingless expression patterns requires the proper activation of both autocrine and paracrine Wingless-pathways, as well as the full activation of the Hedgehog pathway. (3) Protein kinase A plays at least two roles through the phosphorylation of Cubitus interruptus, the effector molecule of the Hedgehog signalling pathway. (4) The roles of Sloppy-paired and Naked in the delineation of the engrailed and wingless expression domains are emphasized as being important for segmental boundary formation. Moreover, the application of an original computational method leads to the delineation of a subset of crucial regulatory circuits enabling the coexistence of specific expression states at the cellular level, as well as specific combination of cellular states inter-connected through Wingless and Hedgehog signalling. Finally, the simulation of altered expressions of segment polarity genes leads to results consistent with the published data concerning loss-of-function mutants and ectopic gene expressions. Moreover, this model makes predictions about genetic interactions that can be tested by further experiments.