Lineage compartments in Drosophila

Abstract

Stable, compartment-specific expression of the selector transcription factors is of central importance for compartments, as it regulates both compartmental identity and signaling between compartments. In most cases each imaginal disc primordium straddles the embryonic parasegment boundary and the state of En-Inv expression is inherited from the parasegmental expression of En-Inv in the embryonic epidermis. The A/P subdivision is thus largely regulated by the embryonic segmentation cascade.The dorsal-specific expression of Ap, however, does not appear until the wing disc contains several hundred cells. Recent work suggests that Ap expression is stimulated by Epidermal Growth Factor (EGF) signaling, perhaps via the neuregulin-like signaling protein Vein, which is expressed in a small cluster of cells near the dorsal end of the wing disc. Once Ap expression is established, however, it becomes insensitive to this signal.And yet, even at later stages of development, selector gene expression can be altered, and the lineage restriction can be changed. For instance, during the regeneration of disc tissue that has been surgically or genetically removed, the compartmental lineage restrictions are lost and then reformed. This is puzzling, as it means that the stable inheritance of the embryonic state of selector expression is not the only way of establishing compartmental domains within the disc. Does this also mean that a regenerating imaginal disc can recapitulate the patterning that occurs within the much smaller embryonic segment? Perhaps, although it is unclear how the gap and pair-rule genes of the segmentation cascade could work in the quite different cellular context of an imaginal disc. However, the existence of compartmental cell affinities provides a possible alternative explanation. During regeneration cells might randomly lose or gain selector gene expression, and thus compartment-specific and boundary-specific cell affinities. As long as this occurred within a small enough region, cells that had regained similar affinities might be able to sort together, reforming the two compartments. After cell proliferation the result would look like spatially patterned gene expression, but in fact would result from random gene expression and regulated cell affinities. Or to put it another way, the same mechanisms that maintain the spatial coherence of compartments during normal development might just be capable of reforming compartments under duress.