Concentration-dependent cellular responses are often induced by the localized production of secreted proteins. One such protein — Hedgehog (Hh) — stimulates responses in distant cells, yet it is expected to remain in the membrane by virtue of its cholesterol and palmitoyl modifications. An insight into how this is possible comes from work published in Cell, in which DispA is identified as an essential factor for releasing Hh.

The authors first identified and characterized two murine homologues of Drosophila disp , mDispA and mDispB, which they propose encode 12-membrane-spanning proteins. On the basis of the early embryonic expression (E7.5–E9.5) of mDispA, and its ability to rescue a disp mutation, the authors studied the role of mDispA in early mammalian Hh signalling by deleting the exon encoding 11 of the 12 transmembrane domains. Homozygous embryos died at around E9.5, showing several patterning defects.

So Ma et al. further studied mDispA's function using fibroblasts from mDispA−/− embryos. A reporter construct containing binding sites for Gli — a transcription factor that is a downstream target of the Hh pathway — showed that mDispA−/− cells responded normally to exogenous Sonic Hh (Shh). This indicated that the mDispA−/− phenotype could be due to the failure either to modify Shh or to present or release it to neighbouring cells.

To test the former possibility, Ma et al. transfected constructs encoding full-length Shh (to which both palmitoyl and cholesterol can be added) or ShhN (which cannot be modified by cholesterol) into mDispA−/− cells and compared the electrophoretic mobility of the resultant proteins. Shh was efficiently cleaved and showed a higher mobility than ShhN, so the lack of Shh signalling in mDispA−/− embryos can't be due to defective Shh processing.

To confirm that the signalling defect arises from the failure to release or present Shh, ShhN-expressing mDispA−/− cells were transfected with mDispA or mDispB expression constructs and mixed with cells containing a Gli reporter. mDispA, but not mDispB, increased the amount of ShhN available for signalling to the reporter cells. Further experiments measuring the release of Shh showed that DispA functions to set Hh proteins free from cell membranes.

The authors propose that Disp proteins have a relatively simple function in initiating communication by catalysing signal release. On the basis of sequence similarity and subsequent mutational analysis, they have shown that Disp proteins are members of the bacterial RND family of transporters, which now has a new function — in membrane release of Hh.