Eukaryote high-level evolution - have we succeeded in mapping it?

[lpetrich]

The overall phylogeny of eukaryotes has long been a difficult and contentious subject, almost as bad as the phylogeny of prokaryotes. One could recognize some well-defined groups, but that was about it.

But as biologists learned out to sequence proteins, and then nucleic acids, they got a solution. Compare molecules that do the same thing in different species and find out how much they differ. One can make a phylogeny or family tree from some molecules and compare them to what was derived using more traditional means. So by the 1970's, biologists started extending their sequencing to organisms with poorly understood phylogenies.

Carl Woese and others discovered, among other things, a deep split in the prokaryotes, the split between (Eu)bacteria and Archaea. Some traditional taxa among the prokaryotes turned out to be well-defined, like cyanobacteria and actinobacteria, but others didn't. They used small-subunit ribosomal RNA, but that molecule's trees have generally been supported by other molecules' trees.

For eukaryotes, the SSU rRNA tree yielded a reasonable-looking result: a crown group of animals, plants, and fungi, and lots of earlier-branching protists, including anaerobic ones. (Molecular Phylogeny first picture)

But there was a problem: other molecules didn't give quite the same tree. What was going wrong? Relatively long branches were discovered to cause discrepancies, among other things.

By the late 1990's, biologists were coming up with rather a very different-looking phylogeny (second picture in the previous link, from http://www.jstor.org/discover/10.1086/303292?uid=3739856&uid=2134&uid=2&uid=70&uid=4&uid=3739256, also see A Kingdom-Level Phylogeny of Eukaryotes Based on Combined Protein Data (open access)). That one used some proteins: elongation factor 1-alpha, actin, alpha-tubulin, and beta-tubulin.

That crown group is gone. Animals and fungi turn out to be relatively closely related, with plants about as distant from them as most other eukaryotes are.Now the present.

Eukaryotes - Tree of Life project (consensus tree)
BMC Biology | Full text | The new micro-kingdoms of eukaryotes
etc.

It's rather close to those 1999 and 2000 results, especially the latter. Here is a summary:

• Unikonts: opisthokonts (animals, fungi), amoebozoans (familiar amoebas, slime molds)
• Archaeplastida: ((green algae / land plants, red algae), glaucophytes)
• SAR: ((Stramenopiles, Alveolata (dinoflagellates, apicomplexans, ciliates)), Rhizaria)
• Excavata: euglenids, diplomonads (Giardia, ...), ...
• Lots of organisms that are not very close to any of these groups or to each other: the "microkingdoms"

Is it too optimistic to expect this result to last? So far, it has held up when one uses as many as a hundred genes or more.

 

[jim mcnamara]

http://tolweb.org/Eukaryotes You can check the articles link on the page to see updated results. Just saw this question today. @Ygggdrasil may have better information.

 

[lpetrich]

Those links are rather out-of-date -- some of them are broken, others don't look like they have been updated recently. So I went over the Google Scholar to get the latest word. It's rather confusing, since a lot of the earlier branchings are difficult to resolve, with some organisms jumping around quite a bit from paper to paper. But here is what seems to be a consensus for the larger groups:

Unikonta / Amorphea / Opimoda contains Opisthokonta and Amoebozoa

Bikonta / Diaphoretickes / Diphoda contains Archaeplastida, SAR, and haptophytes and cryptophytes.

Excavata is split into Discoba (Discicristata) and Metamonada in some recent work. with some papers placing Discoba alongside Diaphoretickes as Diphoda.

So it may be a little bit more until the dust settles. Part of that will involve finding groups with distinctive phenotypes, and that seems to be lacking for some of the larger groups, groups like SAR.