“We are still left hungry for the ancestor of mitochondria,” says Puri Lopez-Garcia, a biologist at the University of Paris-South who was not involved in the study.
While it’s generally agreed that Alphaproteobacteria includes the closest bacterial relatives of mitochondria, that relationship doesn’t reveal much about how mitochondrial ancestors made a living or how they made the jump to acting as organelles. That’s because “Alphaproteobacteria is a particularly diverse group of organisms in terms of kinds of metabolism,” Lopez-Garcia explains. “You find more or less everything in there.” Some studies have found genetic similarities between mitochondria and an order of alphaproteobacterial symbionts known as Rickettsiales, but other, free-living candidates have also emerged.
I do find myself with as many questions at the end of reading the paper as I did at the beginning of reading it, probably just different questions.—James McInerney,
University of Nottingham
The question of where on the alphaproteobacteria family tree the mitochondrial ancestor fell has pestered study coauthor Thijs Ettema throughout his scientific career. “Now, with all the available data from all these new lineages in all sorts of environments, we thought we should just do one bold approach and see where this ends up,” says Ettema, an evolutionary biologist at Uppsala University in Sweden.
Much of the genomic data he and colleagues used in their analysis came from the Tara Oceans dataset, which includes metagenomic sequences from microbes in ocean waters sampled from various depths. “For reasons that are not extremely clear . . . it seems that oceanic waters are extremely enriched for Alphaproteobacteria, and not just one species—it seems to be a whole array,” Ettema explains. The datasets were “good and deep enough to make an effort to reconstruct near-complete genomes.”
“They’ve done their analyses well,” says evolutionary biologist Andrew Roger of Dalhousie University in Halifax, Canada, who was not involved in the study. For example, the researchers accounted for the fact that endosymbionts’ genomes tend to get paired down over time, he explains, and they may lose DNA repair enzymes that fix copying errors that put in an A or a T for another nucleotide. If a substitution falls in a spot where it doesn’t change the amino acid sequence coded for by the gene, it’s likely to stick around, and thus the genomes of mitochondria or organisms such as Rickettsiales grow more AT-rich over time—which can make them seem more closely related than they really are.
One step the research team took to counter that bias was to compare not just the genes of Alphaproteobacteria and mitochondria, but the amino acid sequences of the proteins they produced. Indeed, the investigators found no close relationship between mitochondria and Rickettsiales, or, for that matter, any of the other sampled microbes. “Rather, our analyses indicate that mitochondria evolved from a proteobacterial lineage that branched off before the divergence of all sampled alphaproteobacteria,” they write in their paper.
Another researcher left unsatisfied by the results is Ettema himself. His group plans to continue the search for the long-lost relatives of mitochondria by analyzing samples from non-ocean environments. “It might be that they’re extinct and we’re looking for something that might no longer exist,” he concedes. “Personally, I’m convinced that we’ll find something that will, in the end, be more informative about the origin of mitochondria.”
J. Martijin et al., “Deep mitochondrial origin outside the sampled alphaproteobacteria,” Nature, doi:10.1038/s41586-018-0059-5, 2018.