Higher Plant Proteins of Cyanobacterial Origin: Are They or Are They Not Preferentially Targeted to Chloroplasts?

Dear Editor, What does the evolutionary origin of a plant protein tell about its subcellular localization? Naively thinking, one would assume that plant proteins that were originally encoded in the endosymbiont genome are targeted to the chloroplast. However, published data seem to support only a loose link between evolutionary origin and subcellular localization. About half of the Arabidopsis proteins with a detectable cyanobacterial ortholog are targeted to subcellular compartments other than the chloroplast (Martin et al., 2002). Here we show that the naive view is valid when considering the full phylogenetic profile of plant genes with cyanobacterial orthologs. Genes that are present also in non-photosynthesizing lineages presumably trace back to a primordial eukaryote. They have been inherited largely vertically and show no evidence for a preferential chloroplast targeting. In contrast, genes that are among eukaryotes confined to lineages that have undergone primary or secondary endosymbiosis are likely to be of true cyanobacterial origin. They are indeed mostly targeted to chloroplasts in plants. Unraveling the composition of the chloroplast proteome is crucial for understanding the function and integration of this organelle into the metabolic network of photosynthesizing organisms. Besides photosynthesis, chloroplasts play essential roles in the biosynthesis of amino acids and vitamins, lipids and isoprenoids, the storage of fixed carbon, and other processes. Thus, there is a strong need to tightly coordinate chloroplast activities with the overall metabolism of the cell, and accordingly different retrograde chloroplast-to-nucleus signaling pathways have evolved (Jarvis and Lopez-Juez, 2013). During evolution—after the initial uptake of a cyanobacterium by a heterotrophic host—the cyanobacterium evolved into the contemporary plastid and most of the originally cyanobacterial genes were transferred into the nuclear genome of the host (Martin et al., 2002). Current estimates suggest that 4300–4500 Arabidopsis proteins were acquired from the ancestral plastid. This genetic reorganization created the necessity to establish an effective ‘back-transport’ of the encoded proteins to their original location using an N-terminal signal sequence called cTP (chloroplast transit peptide) (Jarvis and Lopez-Juez, 2013). Technological advances in high-throughput genome sequencing and proteomics have boosted the analysis of the chloroplast proteome. To date, the curated reference plastid proteomes for maize and Arabidopsis (http://ppdb. tc.cornell.edu) comprise 1564 and 1559 proteins, respectively. These numbers are contrasted by those obtained from bioinformatics analysis of the sequenced genomes. About twice as many proteins in these species carry a cTP. This already suggests that a considerable number of chloroplast proteins still remain to be discovered. Unfortunately, the presence of a cTP provides only ambiguous evidence to infer chloroplast localization. For example, of 1325 experimentally identified chloroplast proteins in Arabidopsis, 14% lack an identifiable cTP at their N-terminus (Zybailov et al., 2008). Extrapolations from systematic studies revealed that the fraction of chloroplast proteins that lack a cTP could be about 11% of the total chloroplast proteome (Armbruster et al., 2009). In turn, targeting prediction algorithms, such as TargetP (www.cbs.dtu/dk/services/ TargetP), do not consider N-terminal protein acylation, which can overwrite chloroplast targeting signals resulting in deviating subcellular localizations (Stael et al., 2011). Thus, alternative approaches need to be sought to complement existing information in the prediction of chloroplast targeting. Directed experimental approaches focusing on signaling components had only limited success (Bayer et al., 2011, 2012), possibly due to the low abundance of such proteins in the chloroplast. Considering the evolutionary origin of plant proteins has also been proposed to help predicting their subcellular localization. An orthogenomics approach using 17 species identified 56 Arabidopsis proteins of endosymbiotic origin of which 54 were targeted to chloroplasts (Ishikawa et al., 2009). However, the small number of analyzed proteins makes this finding hard to generalize. Moreover, it