One of the basic developmental mechanisms for patterning a field of cells is to apply a concentration gradient of a morphogen across the field. Depending on its position in the gradient, a cell will experience a certain level of the morphogen, and will respond differently from its neighbours. The superimposition of morphogen gradients in different directions provides positional information that will lead to proper patterning of adult structures in three dimensions (belly–back, bottom–top and proximal–distal). Morphogens such as wingless, decapentaplegic (DPP — a member of the large TGF-β family) and hedgehog are among the biggest stars of developmental biology, but there is an ongoing debate about how gradients of these morphogens are formed. Two complementary studies in Drosophila wing development have provided important new information about the formation of the DPP morphogen gradient.

The adult wing of Drosophila develops from a specific imaginal disc — one of several groups of cells set aside during larval development that give rise to adult structures. The dpp gene is expressed in a central stripe in the wing disc and produces a gradient of DPP signalling away from the stripe, in both directions. The two new studies, by Entchev et al., and Teleman and Cohen, have used a version of DPP tagged with GFP and show, for the first time, that DPP protein is indeed present in a concentration gradient — high in the stripe and low at the edges of the disc. Using different genetic tricks, both studies also show that the DPP gradient can form rapidly, and that gradient formation must be regulated during development. At that point, the two studies diverge.

Entchev et al. show that the formation of the DPP gradient does not occur by passive diffusion outwards from the central stripe. Instead, a variety of mutants that affect endocytosis and intracellular vesicle transport, including shi ts1 , show that DPP is internalized by endocytosis, and is then either degraded or recycled and transported out of the cell. A certain amount of DPP is degraded as it passes through each cell, so the concentration of DPP diminishes with distance from the source. The relative levels of DPP degradation and recycling will influence the shape of the DPP gradient and, as Entchev et al. point out, this provides a means to alter the shape and size of the resultant adult structure.

Teleman and Cohen investigate directly the relationship between the DPP gradient and the size of the cellular field it has to pattern. If cells are growing rapidly, the DPP gradient will have a bigger area to pattern, but will the gradient be able to form over that entire field, to pattern it correctly? Using mutants of the insulin signalling pathway, Teleman and Cohen induced the imaginal disc to grow faster (or slower) on one side of the DPP stripe and at the normal rate on the other side. The end result was that the DPP gradient accommodated these changes, so that the slope of the gradient varied depending on the size of the cellular field. Teleman and Cohen propose that this might be explained by the presence of a DPP sink at the edge of the disc.

Together, these studies provide detailed insight into the way a morphogenetic gradient is formed and how pattern formation by morphogens is coordinated with growth. Given the conservation of the morphogens and their signalling pathways, these insights will have very general relevance for understanding morphogenetic gradients in vertebrate development.