The accepted model for protein import into organelles is that targeting is mediated by receptors that recognize signal sequences. Once cargo proteins have reached the correct membrane, they unfold, thread through the translocon and refold in the lumen of the organelle. Of course, nothing is ever simple, and Dammai and Subramani now report in Cell that Pex5 — the receptor for the peroxisomal targeting signal 1 — actually cycles in and out of peroxisomes.

The authors constructed chimaeras consisting of a PTS2 signal sequence, a prethiolase-processing site, a FLAG epitope and Pex5. The FLAG epitope can be detected by two antibodies: M2, which recognizes it in any context; and M1, which recognizes it only when it is amino terminal — as, for example, after cleavage of the construct by the prethiolase protease. Because the protease cleaves prethiolase only in the peroxisome matrix, cleavage of the construct is diagnostic of entry into peroxisomes.

After expression in HeLa cells, the chimaera was cleaved, indicating that Pex5 had entered peroxisomes. Pex5 was also exported from peroxisomes, as the processed form was largely cytosolic. Peroxisome-associated, processed Pex5 was only partly protected from protease digestion, indicating that it was no longer inside the peroxisome but had docked again to the organelle membrane. Last, kinetic data indicated that processed Pex5 could re-enter peroxisomes, as the appearance of the peroxisomal, processed form accounted for the disappearance of the processed and unprocessed forms from the cytosol.

So peroxisomal import seems to be diametrically opposed to the principles derived from import into mitochondria. Although it remains to be shown that cycling of Pex5 between the cytosol and the peroxisome is relevant for cargo transport, this shuttling mechanism is reminiscent of nuclear transport. The directionality of nuclear transport depends on the GTPase cycle of Ran. So, is there a similar mechanism for the dissociation of Pex5 cargo in the peroxisomal matrix? On the other hand, peroxisomal import resembles the ΔpH pathway of chloroplast thylakoids, in that a fully folded protein (which a receptor would necessarily be) can cross the membrane of an otherwise tightly sealed compartment. Whatever the new mechanism, the paradigm's lost.