Origins

We are, fundamentally, a fusion. As I wrote in my essay for Science on the origin of eukaryotes, there's now a wealth of evidence that our cells evolved from the combination of two different microbes. The mitochondria that generate fuel for our cells started out as free-living bacteria. Today, they still retain traces of their origin in the bacterial DNA they carry, as well as their bacterial structure, including the membrane within a membrane that envelops them.

Scientists I spoke with as I worked on the essay agreed that this merging was a profound event in the history of life. No living eukaryote, whether animal, plant, fungus, or protozoan, has completely lost its mitochondria since that symbiotic milestone some 2 billion years ago. It wasn't the only time that two species merged, however. Plants, for example, descend from algae that engulfed a species of photosynthesizing bacteria. Many protozoans have swallowed up photosynthetic partners as well.

Yet in all these cases, eukaryotes did the swallowing. It's striking that scientists have such a hard time finding an example of a noneukaryote (a prokaryote such as Escherichia coli and other bacteria) hosting a prokaryote symbiont. Some scientists have gone so far as to argue that swallowing up a partner requires lots of intricate molecular systems that can create a pocket in the surface of a cell and can draw that pocket inside the cell as a bubble. Eukaryotes have this sort of cellular skeleton, and prokaryotes, it seems, don't. If that's true, then our ancestors swallowed up mitochondria only after they evolved the molecules necessary for the swallowing.

But today, there's a provocative new alternative to consider. Maybe a lot of today's prokaryotes are also the result of an ancient merger. The idea comes from James Lake of the University of California, Los Angeles, a veteran researcher on the early history of life. In my essay, I describe how Lake first proposed in the early 1980s that the host cell that gave rise to eukaryotes belonged to a lineage of prokaryotes he dubbed eocytes. Now, a quarter of a century later, new studies on genomes are strongly supporting his eocyte hypothesis. In today's issue of Nature, Lake questions whether we may be too quick to assume that only eukaryotes are the result of fusion. He observes that aphids depend on a species of bacterium called Buchnera to digest their food, and Buchnera in turn contains other bacteria on which its own survival depends. These two bacteria are still distinct enough from each other that we can tell them apart. But what if two bacteria joined together billions of years ago and their identities blurred together? How would we tell them apart?

To look for possible signs of ancient fusion, Lake compared proteins in over 3000 different prokaryote genomes. He concluded that a major group of bacteria known as Gram-negative bacteria is actually the result of a fusion of two different kinds of bacteria, known as Actinobacteria and Clostridia. These bacteria, which include the ancestors of mitochondria, are unusual in many ways, but the most obvious one is their membranes. Whereas other bacteria are surrounded by a single membrane, Gram-negative bacteria are surrounded by two. It's possible, Lake argues, that the double-membrane structure of these bacteria is a vestige of one kind of bacteria living inside another.

How exactly they made that merger is one of many questions Lake's hypothesis raises. Buchnera's microbial residents may offer some clues. There are also predatory bacteria that push their way into other bacteria in order to feed on them from the inside; in some cases, these predators spare their victims and just live harmless inside them. Lake also points out that microbes don't actually have to take up residence with each other to mix their genomes.

When scientists dredge up muck from the ocean floor, for example, they often find different species of microbes living together in tight clumps. They have to live close to other species to survive because each species takes care of chemical reactions that their partners can't carry out on their own. That intimacy makes it easier for individual genes to move from host to host, as viruses infect different microbes or as microbes die and other microbes slurp up their genes.

It will be interesting to see if Lake's new hypothesis fares as well as his eocyte hypothesis is doing. If he's right, this symbiosis had an impact on the history of life on par with the origin of eukaryotes. Gram-negative bacteria were the first photosynthesizers, for example, and were then swallowed up by the ancestors of plants. And the same lineage also gave rise to the bacteria that became our own mitochondria. Our cells, in other words, are not just microbes within microbes; they are microbes within microbes within microbes: a true Russian doll of evolution.

—Carl Zimmer