Pathogen to powerhouse

See allHide authors and affiliations

Science  12 Feb 2016:
Vol. 351, Issue 6274, pp. 659-660
DOI: 10.1126/science.aad8864

Mitochondria and plastids are essential for harnessing energy in eukaryotic cells. They are believed to have formed through primary endosymbioses, in which bacterial symbionts were converted into energy-producing organelles. Primary endosymbiosis is extremely rare: Only one other case is known, in the amoeba Paulinella (1). This rarity is usually attributed to the many innovations that are required for organelles to be integrated into the cellular machinery (2). However, the first challenges for an endosymbiont are to avoid being digested by the host and to replicate in its novel environment. Recent studies provide clues to how the precursors to mitochondria and the plastid overcame these challenges.

Molecular phylogenies of mitochondrial genes suggest that the ancestor of the mitochondrion may have been a member of the Rickettsiales, an order of small Alpha-proteobacteria (see the figure) (3). These intracellular organisms are often referred to as energy parasites because they encode a protein responsible for the unidirectional import of adenosine 5′-triphosphate from the host cytosol. Specialized Rickettsiales pathogens transmitted by arthropods have a highly reduced gene inventory (4). However, other Rickettsia pathogens have relatively large genomes and appear to infect various protists (5). Unlike the animal pathogens, these intracellular Rickettsiales are deeply rooted within Alphaproteobacteria (3) and are therefore more likely to resemble the mitochondrial ancestor.

Research also provides clues to a possible host for the mitochondrial endosymbiosis. A recently discovered archaeal lineage, the Lokiarchaeota (see the figure), is unique among Archaea in that it shares an array of signature eukaryotic genes thought to be critical for enabling endosymbiosis. These genes are involved in membrane remodeling, vesicular trafficking, and endocytosis. The latter would have allowed the capture of the mitochondrial precursor (6). The discovery of the Lokiarchaeota thus provides a potential missing link in the story of eukaryote evolution: membrane trafficking, mitochondrion-lacking cells that could have hosted the mitochondrial endosymbiosis.

It is likely that endocytosis in Archaea originally evolved to facilitate the harvesting of macromolecules in nutrient-rich external media. This process may have opened the door to incidental capture of harmful bacteria. The resulting intimate contact between the host and foreign cells likely precipitated an evolutionary arms race: The host evolved mechanisms such as autophagy, production of antimicrobial peptides, and reactive oxygen species induction to protect itself (7), whereas foreign cells evolved virulence effector molecules to evade host defense mechanisms. Over time, some foreign cells became obligate intracellular pathogens suited to life in the cytosol of the cell lineage that was to become the eukaryotic ancestor. Many effectors evolved to interact specifically with host proteins to allow replication of the intracellular pathogen, thereby preadapting them to symbiosis.

If we assume that this series of events is reasonably accurate, then free-living bacteria lacking the toolkit for intracellular life would not have survived long enough in the well-protected host cell cytosol to give rise to the mitochondrion. It is thus significant that Rickettsia is among the few pathogens that can multiply directly in the host cytosol (rather than in membrane-bound vacuoles termed inclusion vesicles). It thereby meets a key requirement for endosymbiosis: the ability to withstand host defenses.

Pathogen roots of eukaryotic organelles.

The original archaeal host cell was a member of the Lokiarchaeota that underwent primary endosymbiosis (yellow circle) with a Rickettsiales-like obligate intracellular bacterium, giving rise to the mitochondrion (6). Thereafter, the Archaeplastida ancestor underwent primary endosymbiosis (blue circle) with a cyanobacterium, putatively aided by a chlamydial infection (red circle) (12, 13).


These results also help explain why ancestrally mitochondrion-lacking eukaryotes (so-called archezoans) have never been found: They likely do not exist. The mitochondrion was already present in the Lokiarchaeota ancestor of eukaryotes and powered the transition from endocytosis of nutrients to a full-fledged phagocytic lifestyle (8). And finally, the Lokiarchaeota discovery has shifted focus away from metabolic symbiosis models for eukaryote origin (the hydrogen hypothesis) (9, 10) that rely on cell fusion. The mitochondrion appears to have originated via classical endosymbiosis, whereby a host cell engulfs and maintains a foreign cell.

When we turn our attention to primary plastid origin, the story becomes more complex. The host of this endosymbiosis was a mitochondrion-containing unicellular eukaryote, but the cyanobacterial endosymbiont was likely not a pathogen. Living cyanobacteria neither possess the genetic toolkit to evade host defenses, nor do they encode effector proteins to interact with the host cellular machinery. So, how did the unprotected photosynthetic cell survive the early phases of the endosymbiosis? A potential solution to this conundrum was provided by the discovery of several dozen genes of chlamydial origin in the nuclear genome of algae and plants (11). These results suggest that a chlamydial bacterium entered the host cell together with a cyanobacterium (see the figure). This allowed the cyanobacterium to escape host defenses and establish a tripartite symbiosis through the help of chlamydial-encoded effector proteins and transporters (12, 13). However, the details of this complex process remain incompletely understood.

As the recent studies discussed here show, we are in an exciting phase of endosymbiosis research. However, we still lack some crucial information. The Lokiarchaeota were identified from assembly of metagenomic data (6); no living cells have yet been isolated to validate the genome assembly or test their actual physiological capabilities. Protist-infecting chlamydial cells with large genomes are continuously being isolated, but only a handful have been studied in detail and are represented in genome databases. Therefore, more data are needed to test the proposed role of pathogens in organelle origin.

But perhaps what we need most are attempts at experimental primary endosymbiosis in the lab to learn the rules underlying this remarkable process. These data can be used to test the pathogen-centered view of primary endosymbiosis presented here and may help us get beyond phylogenies and simple diagrams. Ultimately, these advances will help us address the most vexing question about primary endosymbiosis: Why is it so rare?

References and Notes

Acknowledgments: S.G.B. was supported by the CNRS, the Université des Sciences et technologies de Lille, and the ANR grants “expendo” and “ménage à trois.” D.B. was supported by NSF grants MGSP 0625440 and MCB 0946528, and A.P.M.W. was supported by German Research Foundation grants CRC-TR1, CRC 1208, and EXC 1028. All authors contributed equally to this work. We thank G. Greub, Ch. Colleoni, S. Aebi, M. Ducatez, B. Bouchet, H. Qiu, and D. C. Price for helpful comments on the manuscript. In memory of Fabrice Rappaport, a distinguished student of organelle powerhouses.

Navigate This Article