Curr Genet (1991) 19:503-507 Current Genetics 9 Springer-Verlag 1991 Evidence for multiple xenogenous origins of plastids: comparison of psbA-genes with a xanthophyte sequence Siegfried Scherer *, Gerhard Herrmann 1, Joseph Hirschberg 2, and Peter Bfger * 1 Lehrstuhl ffir Physiologieund Biochemieder Pflanzen, Universit/it Konstanz, GieBberg,W-7750 Konstanz, Federal Republic of Germany z Department of Genetics, The Hebrew University, Jerusalem 91904, Israel Received January 10/February 21, 1991 Summary. When only plastidic features are considered, it is difficult to distinguish between monophyletic and poly- phyletic xenogenous origins of plastids. We suggest that a direct comparison of nuclear and plastidic sequence- similarity pattern will help to solve this problem. The D1 amino acid sequence of six major groups of photosyn- thetic eukaryotes and of the two groups of photosynthet- ic prokaryotes are now available, including the psbA- gene product from Bumilleriopsis filiformis, which is the first molecular sequence reported for a xanthophycean alga. Evidence is provided for an independent and poly- phyletic origin of plastids from five out of the six major taxa of photosynthetic eukaryotes. This conclusion is reached by comparing a plastid-based pattern of D1 sim- ilarity with a nucleus-based similarity pattern published recently. Furthermore, the availability of D1 sequences from five eukaryotic algae led to a re-evaluation of the taxonomic position of Prochlorothrix. Key words: Endosymbiosis -psbA - Xanthophyceae - Prochlorothrix, Bumilleriopsis - Eukaryotic algae Introduction Molecular data have played an instrumental role in the discussion of whether plastids are of autogenous or xenogenous origin (Giovanni et al. 1988). It is known that not only molecular sequences (Gray and Doolittle 1982; Giovanni et al. 1988) but also gene organization and mode of expression, as well as other features of plas- tids, closely resemble those of eubacteria (Margulis 1981; Golden et al. 1986; Cavalier-Smith 1987) and are funda- mentally different from those of the nucleus. Conse- quently, the xenogenous origin of plastids is no longer a matter for discussion. * Current address: Bakteriologisches Institut, Technische Univer- sit/it M/inchen, V6ttingerstrasse 45, W-8050 Freising, Federal Re- public of Germany Offprint requests to." P. B6ger The classification of photosynthetic eukaryotes is dis- puted. The differences in pigment composition and ultra- structure of the plastid, as well as features of the nucleus and flagella, led to the separation of at least four, but more often seven to ten, major groups. For instance, the superkingdom of eukaryotes may be divided into nine kingdoms. Within these, five kingdoms include photo- synthetic organisms (Cavalier-Smith 1981; Gray 1989). Considering the vast differences between these five king- doms, the general proposal of a multiple polyphyletic origin of plastids (Margulis 1981; Raven 1970; Whatley and Whatley 1981) seems understandable, but is never- theless actively disputed (Cavalier-Smith 1982, 1987; Kowallik 1989). Cavalier-Smith (1982), for instance, em- phatically suggests a monophyletic origin of all plastids, interpreting the differences by subsequent loss or conver- gent evolution of characters, whereas the nucleotide se- quences ofplastidic genes coding for ribulose-l,5-bispho- sphate carboxylase/oxygenase in Cyanophora (Valentin and Zetsche 1990a), Antithamnion (Kostrzewa et al. 1990) and Cyanidium (Valentin and Zetsche 1990b) have been discussed in a polyphyletic model. To contribute to this vivid discussion, suitable se- quence data from as many photosynthetic eukaryotes as possible must be gathered from both plastid DNA and nuclear DNA. We suggest that the comparison of nuclear with plastidic sequence-similarity pattern will help to dis- tinguish more clearly between monophyletic and poly- phyletic origins of plastids. Such a comparison has not yet been performed. Accordingly, we decided to compare nuclear sequence-similarity patterns which have been published recently (Perasso et al. 1989; Sogin et al. 1989) with that of the psbA gene, coding for the Dl-protein of photosystem II (herbicide-binding protein). This is a most suitable probe for such evaluations since it is local- ized on the plastome of all eukaryotic algae and plants so far examined, is ubiquitous for oxygen-evolving photo- systems, highly conserved and comparatively long (360 amino acids), all of which is essential when deeply branched groups of protists have to be resolved.