The juvenile salivary glands (green) stick around in this pupa with a mutation in the CBP gene.

As it morphs from a squirming maggot into a buzzing adult, a fruit fly jettisons its juvenile salivary glands. On page 85, Yin et al. show how the insects get rid of a protective protein, allowing the glands to break down at the right time. The research helps clarify how certain tissues prepare themselves to die.

During metamorphosis, a larva's obsolete internal organs degenerate, and replacements sprout. Surging quantities of the steroid hormone ecdysone spur the salivary glands and other juvenile structures to melt away. Researchers have mapped out some of the molecular details of this deterioration. Ecdysone cranks up several genes, including one called reaper, that trigger salivary gland cells to perish. But getting rid of the glands requires one more change: elimination of a protein called diap1, which shields salivary cells. How a fly-to-be cuts the levels of diap1 on schedule was unknown.

To find out, the researchers screened the insects for mutations that allowed the youthful salivary glands to persist. They found three such glitches in the transcriptional coactivator CBP. The scientists assumed that faulty CBP spared the salivary glands by preventing the activation of reaper and other killer genes. Instead, Yin et al. determined, CBP slashed the amount of diap1 shortly before pupation.

Although the decline in diap1 primed the salivary glands to die, enough of the protein remained to spare cells from reaper. A second surge of ecdysone triggered salivary cells to purge the vestiges of diap1 without using CBP, and the glands deteriorated. This two-step process ensures that the salivary glands don't disappear prematurely, the researchers speculate. They now want to pin down how CBP influences diap1. CBP normally helps switch genes on rather than off, so it probably works by activating another gene that curtails diap1 production.