Eukaryotic Origins: Revolution in the Classification of Life

Peer through a microscope at some pond water today and you might see what Antonie van Leeuwehnoek or Robert Hooke saw in the mid-1600s: tiny organisms darting, wiggling and oozing to and fro. These single-celled organisms come in two distinct sizes, the larger ones easily big enough to engulf the smaller ones.

Now it’s a new millennium and scientists are still struggling to understand the evolutionary relationships among different types of cells. The small cells van Leeuwenhoek saw – mere specks even at 300x magnification – are the Archaea and the Bacteria.

mitochondrion_and_chloroplast

Eukaryotes as a group also contain other so-called organelles, such as mitochondria (left) and chloroplasts (right).

The larger cells swimming in pond water are essentially similar to the cells of all multicellular organisms, including humans. In the 1820s, as scientists began to revive the microscopic study of cells after a century-long hiatus, they discovered that cells of larger single-celled organisms, plants, fungi and animals all contain a large more or less central part which absorbs certain stains. They called this the nucleus, and cells that contain nuclei eukaryotes (true-kernel cells). The most recent classification of all life on Earth, then, includes three domains: Archaea, Bacteria (also called Eubacteria) and Eukarya, each of which contains a number of kingdoms.

How?

Eukaryotes as a group also contain other so-called organelles, such as mitochondria and chloroplasts. Less than 100 years after the discovery of the nucleus, a Russian scientist named Mereschowsky first suggested that at least some organelles, including the nucleus, evolved from bacteria engulfed by early cells.

Most scientists today accept that at least the mitochondria (seen in nearly all eukaryotes) and chloroplasts (seen in all photosynthetic eukaryotes) descended from engulfed bacteria. Whether the nucleus resulted from such an endosymbiotic event remains hotly debated.

Hyman Hartman, an evolutionary biologist at MIT, weighs in in favor of the nucleus being an endosymbiont. "Let’s put it this way," Hartman says. "It’s obviously an interesting hypothesis but not the fashionable one. I’m trying to revive it and I think that I am at this time the foremost advocate for it."

Hartman’s latest attempt, a February 2002 paper published in Proceedings of the National Academy of Sciences, also asks another question: Who engulfed whom?

Engulfing’s not an easy task. Archaeans and bacteria can’t do it. To "eat," they generally absorb nutrients from their surroundings, sometimes secreting digestive enzymes to break down more-complex environmental substances. They have none of the complex cellular apparatus, the cytoskeleton, that makes it possible for a cell to wrap part of itself around another cell. (At the May meeting of the Society for Microbiology, however, Cheryl Jenkins and James T. Staley, of the University of Washington, presented a poster reporting that the bacterium Prostehcobacter dejongeii has two genes remarkably similar to cytoskeletal genes of eukaryotes.)

Hartman proposes a separate ancestral cell type, which he calls a chronocyte. He suggests that a chronocyte (C) engulfed both archaeans (A) and bacteria (B) to form eukaryotes (E): E = A + B + C, making today’s eukaryotes a combination of all three cell types.

"There’s no doubt that there’s been a B because the mitochondrion and chloroplast are coming from the eubacteria. And there’s no doubt that there was an archaean because in some basic sense a lot of the informational proteins are coming from archaeans. So C + A is not good enough or C + B is not good enough," Hartman says.

jere_lipps
When Jere Lipps, a paleontologist at the University of California, Berkeley, adjusted a tree showing the emergence of eukaryotes to remove long-branch attraction artifacts, the "tree" turned into a bush with a long stem, with all eukaryotes emerging in a geologically short period of time.
Credit: UC Berkeley

In the PNAS paper, Hartman presents evidence for his ABC hypothesis: a unique set of eukaryotic proteins not found in any bacterium or archaean. Hartman found 347 proteins he and coauthor Alexei Federov, of Harvard, call eukaryotic signature proteins. Among these proteins are several associated with the cytoskeleton, leading Hartman to the conclusion that the chronocyte had the mechanisms necessary to engulf other cells.

Clues gleaned from this protein set and other research lead Hartman to the conclusion that the chronocyte stored its genetic information in RNA rather than DNA. This further supports his contention that the eukaryotic nucleus, a DNA-based structure, arose from a symbiosis of the chronocyte and a DNA-based archaean.

This symbiosis moved genes from one type of cell directly to another type, an example of horizontal gene transfer. Evidence suggests that the symbiotic event in which chronocytes captured both archaeans and bacteria happened around 2 billion years ago "And that’s precisely when oxygen comes into the atmosphere." Hartman says. While we thrive in an oxygen-rich atmosphere, the gas was poisonous to many of the organisms on Earth when photosynthesis arose. "It forced cells into horizontal transfer due to the fact that they had to react to this new poison, oxygen," he says.

When?

The advent of molecular biology began a revolution in the classification of organisms. With the new tools for looking at molecules, biologists began attempting to figure out relationships between organisms based not on how they looked, but on the sequences of subunits in their proteins and nucleic acids. A new tree of life emerged from the work of Carl Woese, at the University of Illinois at Urbana-Champaign, and others.

But as with any new technology, molecular phylogeny requires time to mature. Woese’s early work relied on a single molecule, an RNA found in the protein synthesis machinery of the cell. And it required computing power.

Jere Lipps, a paleontologist at the University of California, Berkeley, looked at the tree diagrams produced by molecular phylogenists and saw a problem. "There’s alwaysbeen a worry about that tree in my mind and in other people’s minds, and that has to do with the fact that at the base of the tree you have Giardia and several other vertebrate-parasite groups. And when I first saw that I thought, ‘Wow, that can’t be right; there’s got to be something wrong with this.’"

The problem arises as computers repeatedly calculate similarities in the sequences of proteins or nucleic acids. Called long-branch attraction, the glitch can make groups seem more related – that is closer together on the tree – than they should be.

When Lipps adjusted a tree showing the emergence of eukaryotes to remove long-branch attraction artifacts, the "tree" turned into a bush with a long stem. Instead of many groups branching off sequentially, followed by a final crown group, the new diagram looks more like a single "star burst," Lipps says, with all eukaryotes emerging in a geologically short period of time.

horizontal_gene_transfer

The concept of horizontal gene transfer (shown above) has clearly moved from a quirk seen in certain organisms to a driving force in evolution. All three domains have genetic material clearly derived from other domains.
Image Credit: Doolittle, 1999

"I wouldn’t call it an explosion," Lipps says, "I’m sure it took place slowly, and that each lineage would have evolved at different rates. It wouldn’t be that all the lineages emerged independently at one time."

Combined with the geological record, this reanalysis suggests that eukaryotes arose earlier than previously supposed, possibly as long as 2.7 billion years ago.

What’s Next?

The concept of horizontal gene transfer has clearly moved from a quirk seen in certain organisms to a driving force in evolution. All three domains have genetic material clearly derived from other domains.

The human genome contains genes from both archaeans and bacteria, and the genome of one organism thought to be primitive, Thermotoga maritima, was found to be one-quarter borrowed from other organisms.

Woese’s original tree clearly needed to be rethought after the Termotoga results were published. "Carl Woese then decided that with this amount of horizontal transfer, he had to now justify what is a domain," Hartman says. "He set out to look for proteins that were unique to the Archaea, and that was what gave me the idea to look for unique proteins in the eukaryotes. (For Woese’s latest theory, suggesting that horizontal gene transfer is even more important than previously thought, see "On the Evolution of Cells," published electronically by PNAS on June 17, 2001.)

Lipps emphasizes that molecular phylogeny remains a work in progress. "I think the molecular phylogenies need to be adjusted, since they give us a single topology based on a single molecule or part of a molecule. People are trying to do this with more comprehensive trees [combining molecular data] with morphology and with the stratigraphy that we know about. So there’s a lot more to do, and all of this has got to be considered extremely tentative. Everything that’s been proposed, in my mind, is a working hypothesis, which is fine, that’s the way science proceeds. I think it’s an exciting time in this evolutionary biology business because of the convergence of paleontology, molecules and traditional sorts of biological approaches."