Phylogenomic Evidence for Separate Acquisition of Plastids in Cryptophytes, Haptophytes, and Stramenopiles Denis Baurain  ,1,2 Henner Brinkmann, 1 Jo ¨rn Petersen, 3 Naiara Rodrı ´guez-Ezpeleta,à ,1 Alexandra Stechmann, 4 Vincent Demoulin, 2 Andrew J. Roger, 4 Gertraud Burger, 1 B. Franz Lang, 1 and Herve ´ Philippe* ,1 1 De ´partement de Biochimie, Centre Robert-Cedergren, Universite ´ de Montre ´al, Montre ´al, Que ´bec, Canada 2 Algologie, Mycologie et Syste ´matique expe ´rimentale, De ´partement des Sciences de la Vie, Universite ´ de Lie `ge, Lie `ge, Belgium 3 DSMZ—Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany 4 Department of Biochemistry and Molecular Biology, Dalhousie University, Halifax, Nova Scotia, Canada  Present address: Unit of Animal Genomics, GIGA-R and Faculty of Veterinary Medicine, Universite ´ de Lie `ge, Lie `ge, Belgium. àPresent address: Functional Genomics Unit, CIC bioGUNE, Derio, Bizkaia, Spain. *Corresponding author: E-mail: herve.philippe@umontreal.ca. Associate editor: Martin Embley Abstract According to the chromalveolate hypothesis (Cavalier-Smith T. 1999. Principles of protein and lipid targeting in secondary symbiogenesis: euglenoid, dinoflagellate, and sporozoan plastid origins and the eukaryote family tree. J Eukaryot Microbiol 46:347–366), the four eukaryotic groups with chlorophyll c–containing plastids originate from a single photosynthetic ancestor, which acquired its plastids by secondary endosymbiosis with a red alga. So far, molecular phylogenies have failed to either support or disprove this view. Here, we devise a phylogenomic falsification of the chromalveolate hypothesis that estimates signal strength across the three genomic compartments: If the four chlorophyll c–containing lineages indeed derive from a single photosynthetic ancestor, then similar amounts of plastid, mitochondrial, and nuclear sequences should allow to recover their monophyly. Our results refute this prediction, with statistical support levels too different to be explained by evolutionary rate variation, phylogenetic artifacts, or endosymbiotic gene transfer. Therefore, we reject the chromalveolate hypothesis as falsified in favor of more complex evolutionary scenarios involving multiple higher order eukaryote–eukaryote endosymbioses. Key words: eukaryote–eukaryote endosymbioses, chromalveolate hypothesis, phylogenomic falsification, variable length bootstrap, multigene analysis. Introduction Bacterial endosymbiosis has been a major evolutionary force in shaping eukaryotic cells as we know them today. In an ancient event, an alphaproteobacterium gave rise to mitochondria, which house respiration and oxidative phos- phorylation along with a variety of other functions (Embley and Martin 2006). A subsequent endosymbiosis with a cyano- bacterium resulted in plastids capable of photosynthesis. These ‘‘primary’’ plastids occur in Plantae (Archaeplastida), that is, green algae and land plants, red algae, and glauco- phytes (Palmer 2003; Reyes-Prieto et al. 2007; Gould et al. 2008). More recently in evolutionary history, nonphotosyn- thetic eukaryotes came across another way of acquiring plas- tids: engulfing a photosynthetic eukaryote instead of a photosynthetic bacterium. When the symbiont is a member of Plantae, the event is termed ‘‘secondary’’ endosymbiosis, while ‘‘tertiary’’ (‘‘quaternary’’ etc.) endosymbiosis designates engulfment of a symbiont that is already the product of a pre- ceding eukaryote–eukaryote endosymbiosis (EEE). Evidence for EEEs comes from the presence of three or four mem- branes surrounding the plastid, which requires targeting of nucleus-encoded plastid proteins with multipartite pre- sequences (Cavalier-Smith 2003), and from less reduced states, where the endosymbiont retains a vestigial nucleus (nucleomorph) (Ludwig and Gibbs 1987). Cryptophytes, alveolates, stramenopiles (heterokonts), and haptophytes (in the following collectively referred to as CASH) are four diverse and ecologically important eukaryotic lineages that include both photosynthetic and nonphotosynthetic taxa. CASH plastids likely arose from a single initial event of secondary endosymbiosis with a red alga (Palmer 2003; Reyes-Prieto et al. 2007; Gould et al. 2008) because 1) all photosynthetic species of CASH have chlorophyll c, which is absent from all other algae (includ- ing reds), and 2) in phylogenies based on plastid-encoded genes (Yoon et al. 2002), as well as on certain nucleus- encoded proteins involved in plastid function, for example, plastid-targeted glyceraldehyde-3-phosphate dehydroge- nase (GAPDH) (Fast et al. 2001), class II fructose-1,6- bisphosphate aldolase (FBA-II) (Patron et al. 2004), and phosphoribulokinase (PRK) (Petersen et al. 2006), a mono- phyletic group of CASH lineages is recovered. Yet, it has been controversial whether or not CASH plastids (and nucleus-encoded gene products functioning in plastids) were inherited strictly vertically, as postulated in the © The Author 2010. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org 1698 Mol. Biol. Evol. 27(7):1698–1709. 2010 doi:10.1093/molbev/msq059 Advance Access publication March 1, 2010 Research article