Plastids and protein targeting

Plastids with two bounding membranes-as exemplified by red algae, green alg ae, plants, and glaucophytes-derive from primary endosymbiosis; a process i nvolving engulfment and retention of a cyanobacterium by a phagotrophic euk aryote. Plastids with more than two bounding membranes (such as those of eu glenoids, dinoflagellates, heterokonts, haptopytes, apicomplexa, cryptomona ds, and chlorarachniophytes) probably arose by secondary endosymbiosis, in which a eukaryotic alga (itself the product of primary endosymbiosis) was e ngulfed and retained by a phagotroph. Secondary endosymbiosis transfers pho tosynthetic capacity into heterotrophic lineages, has apparently occurred n umerous times, and has created several major eukaryotic lineages comprising upwards of 42,600 species. Plastids acquired by secondary endosymbiosis ar e sometimes referred to as "second-hand." Establishment of secondary endosy mbioses has involved transfer of genes from the endosymbiont nucleus to the secondary host nucleus. Limited gene transfer could initially have served to stabilise the endosymbioses, but it is clear that the transfer process h as been extensive, leading in many cases to the complete disappearance of t he endosymbiont nucleus. One consequence of these gene transfers is that ge ne products required in the plastid must be targeted into the organelle acr oss multiple membranes: at least three for stromal proteins in euglenoids a nd dinoflagellates, and across five membranes in the case of thylakoid lume n proteins in plastids with four bounding membranes. Evolution of such targ eting mechanisms was obviously a key step in the successful establishment o f each different secondary endosymbiosis. Analysis of targeted proteins in the various organisms now suggests that a similar system is used by each gr oup. However, rather than interpreting this similarity as evidence of an ho mologous origin, I believe that targeting has evolved convergently by combi ning and recycling existing protein trafficking mechanisms already existing in the endosymbiont and host. Indeed, by analyzing the multiple motifs in targeting sequences of some genes it is possible to infer that they origina ted in the plastid genome, transferred from there into the primary host nuc leus, and subsequently moved into the secondary host nucleus. Thus, each st ep of the targeting process in "second-hand" plastids recapitulates the gen e's previous intracellular transfers.