Secondary endosymbioses and evolution of unicellular eukaryotes

Abstract

Many lines of evidence support the idea that the first chloroplast was the result of an endosymbiotic relationship between a cyanobacterium and a non‐photosynthetic eukaryote. Most of the cyanobacterial genes were lost, but a few remained in the chloroplast genome and as many as a thousand were transferred to the host nucleus. Genes encoding functions required by the chloroplast had to acquire presequences to target the products to the chloroplast. The situation gets more complicated when we consider the algae with chlorophyll c. They are the product of secondary endosymbiosis, where a putative red algal ancestor was engulfed by another non‐photosynthetic eukaryote, which retained the red algal chloroplast but eventually got rid of the rest of the cell. This left the chloroplast surrounded by two additional membranes: one derived from the red algal plasma membrane and the other from the host's phagocytic vacuole. In order for the endosymbiotic relationship to work, there must have been a substantial amount of gene transfer from the red algal nucleus to the host nucleus to support chloroplast functions. In the cryptophytes we even see an intermediate stage in this process, a relict nucleus (nucleomorph) in the periplastidal space between the outer two membranes and the original chloroplast envelope. Now that the draft genome sequence of the diatom Thalassiosira pseudonana (Diatom Genome Consortium) as well as genomes of rhodophyte Cyanidioschyzon merolae and green plants are available, it is possible to investigate the evolutionary history of plastid localized metabolic pathways. Phylogenetic analyses of nuclear‐encoded putatively plastid‐targeted enzymes showed that plastids obviously utilize enzymes not only of expected plastid (cyanobacterial) origin. Within the diatom, apicomplexan, plant and rhodophyte genomes, we have identified several enzymes that originate in α‐proteobacteria (mitochondria) or even in eukaryotic nucleus, but possess N‐terminal plastid‐targeting presequences. Although diatoms are, according to multiprotein phylogeny, related to Alveolates, some plastid‐related metabolic pathways show substantially different evolutionary pattern as well as, in silico, predicted localizations of involved enzymes.