Janouškovec J, Tikhonenkov DV, Burki F, Howe AT, Rohwer FL, Mylnikov AP, Keeling PJ
Curr. Biol. 27 (23) 3717-3724.e5 [2017-12-04; online 2017-11-22]
The origin of eukaryotic cells represents a key transition in cellular evolution and is closely tied to outstanding questions about mitochondrial endosymbiosis [1, 2]. For example, gene-rich mitochondrial genomes are thought to be indicative of an ancient divergence, but this relies on unexamined assumptions about endosymbiont-to-host gene transfer [3-5]. Here, we characterize Ancoracysta twista, a new predatory flagellate that is not closely related to any known lineage in 201-protein phylogenomic trees and has a unique morphology, including a novel type of extrusome (ancoracyst). The Ancoracysta mitochondrion has a gene-rich genome with a coding capacity exceeding that of all other eukaryotes except the distantly related jakobids and Diphylleia, and it uniquely possesses heterologous, nucleus-, and mitochondrion-encoded cytochrome c maturase systems. To comprehensively examine mitochondrial genome reduction, we also assembled mitochondrial genomes from picozoans and colponemids and re-annotated existing mitochondrial genomes using hidden Markov model gene profiles. This revealed over a dozen previously overlooked mitochondrial genes at the level of eukaryotic supergroups. Analysis of trends over evolutionary time demonstrates that gene transfer to the nucleus was non-linear, that it occurred in waves of exponential decrease, and that much of it took place comparatively early, massively independently, and with lineage-specific rates. This process has led to differential gene retention, suggesting that gene-rich mitochondrial genomes are not a product of their early divergence. Parallel transfer of mitochondrial genes and their functional replacement by new nuclear factors are important in models for the origin of eukaryotes, especially as major gaps in our knowledge of eukaryotic diversity at the deepest level remain unfilled.