Deciding what to eat

C ells adapt to their surroundings by generating new organelles when needed and degrading them when they become damaged or superfl uous. Many organelles, including the ER and mitochon-dria, are turned over by autophagy—the " self-eating " pathway in which cytoplasmic contents are engulfed in a double-mem-braned vesicle and transported to the lyso-some to be broken down and recycled. But, just like people, cells have to watch what they eat and ensure only the right organelles are degraded. Manjithaya et al. now reveal the signaling pathway that specifi cally stimulates the destruction of peroxisomes (1). Because peroxisomes contain the enzymes necessary for catabolizing fatty acids, yeast grown on oleate produce extra peroxi-somes to exploit this carbon source. But switching the yeast to glucose-containing medium renders these peroxisomes redundant. In a process called pexophagy, cells dismantle their excess peroxisomes using their autophagic machinery (2). " But how do you take a nonselective process like au-tophagy and make it only eat peroxi-somes? " asks Suresh Subramani, from the Subramani and colleagues have identified factors that control the selective destruction of peroxisomes, including a protein called Atg30 that targets the organ-elles to the autophagy pathway (3). Little is known about the signaling events that control pexophagy, however , so Manjithaya et al. screened for yeast kinase mutants that couldn't eliminate their peroxisomes after switching from oleate to glucose media (1). The researchers discovered that the mitogen-activated protein kinase (MAPK) Slt2p was essential for peroxi-some turnover. " Our assay was pexophagy-specifi c but we were concerned that this might be a general signaling pathway for all autophagic processes, " says Subramani. " We were really excited to fi nd that Slt2p didn't affect general autophagy or the selective removal of other organelles. " Slt2p is at the bottom of a signaling cascade activated by damage to the yeast cell wall (4). All of the upstream kinases in the pathway were also required for pexophagy, as was Mid2p—a cell surface protein that senses cell wall integrity and initiates sig-naling to Slt2p. But what does the cell wall have to do with peroxisome turnover? In fact, activating the MAPK pathway wasn't sufficient to induce pex-ophagy. The team thinks that simultaneous activation of protein kinase A (PKA) by the entry of hexoses, such as glucose, into yeast cells is also required for peroxisome removal. The presence of glucose stimulates PKA by up-regulating cAMP levels and, Subramani speculates, …