The Three Domain Hypothesis (part 5)

[Part 1][Part 2][Part 3][Part 4]
If the Tree of Life Fell, Would We Recognize the Sound?

I've been summarizing a series of papers that appeared in a recent book, Microbial Phylogeny and Evolution. The underlying theme is the validity of the Three Domain Hypothesis. The Three Domain Hypothesis is the idea that life can be divided into three distinct (monophyletic) domains: archaebacteria, bacteria, and eukaryotes. An important part of the hypothesis is that eukaryotes are descended from ancient archaebacteria.

I've been arguing that the Three Domain Hypothesis has been refuted. It is effectively dead even though most people don't realize it. It continues to be touted in the textbooks in spite of the fact that experts have rejected it as an accurate model of early evolution.

Ford Doolittle has been thinking about molecular evolution for over three decades. Some of you might recall that he was one of the originators of the term "selfish gene." He was a strong supporter of the Introns-Early Hypothesis back in the 1980's but he abandoned it when it was falsified by the accumulation of nasty facts. This is one of the reasons why Ford Doolittle is so highly respected in the community of molecular evolutionists.

As usual, Ford Doolittle offers an insightful analysis of the controversy in his paper "If the Tree of Life Fell, Would We Recognize the Sound?" He points out that different genes give different trees at the deepest level and this is a serious problem. One that can't be ignored. Are there some "hard core" genes that have not been transferred during the exchange phase of evolution, where gene swapping was common? If so, these could reveal the "true" deep phylogeny?

His answer is no. Surprisingly, there are only about 100 genes found in all prokaryotic species. There don't appear to be any standard subset that are more reliable than others. All genes are candidates for orthologous replacement by gene transfer from another species. Doolittle makes the obvious, but often ignored, point that you can only determine if a gene has been exchanged when you have a true phylogeny to compare it with. You can't just assume that your favorite genes reveal the true phylogeny and all others conflicting with it are artifacts of lateral gene transfer. This is what disciples of the Three Domain Hypothesis often do. It's called begging the question.

Doolittle thinks this is bad science ...

There may be a real catch-22 in assessing how much LGT as orthologoous replacement afflicts the core at depth, but given that we know that orthologous replacement can happen and that rampant LGT drives genome (gene content) evolution at the strain-in-a-species level, there is no justification in retaining vertical descent as the null hypothesis and requiring stronger proof for LGT. Rather (at least with greater fairness), we might recast the notion of the existence of a stable core as a hypothesis that needs to be tested, not a truth that needs further elaboration. If the hypothesis is that there is a cadre of genes that have never been exchanged (and that thus track organismal phylogeny), and the test of it requires that there indeed be such universally shared genes that show the same phylogeny, then this hypothesis has yet to be proven.

What about the "complexity hypothesis?" This hypothesis refers to a core of genes that have never been transferred from one species to another because they are all part of a large complex. Presumably, the pieces of this complex are not interchangeable so new genes cannot be accepted. Thus, this core represents the true phylogeny of the lineage and all other genes have been acquired later.

The "true" core, according to this argument, is the complex of genes that are involved in translation. This includes ribosomal RNA and ribosomal protein genes.

There are three arguments against the complexity hypothesis. First, genes for some key translational components do not agree with the ribosomal RNA tree, refuting the idea that all genes of the complex evolved together.

Second, Doolittle points out that the logic is flawed. For example, the parts of ribosomal RNA that interact with ribosomal proteins are highly conserved so there is very little difference between species. The parts that don't interact are quite variable and those are the very parts that determine the phylogeny. Lateral gene transfer of ribosomal RNA genes from one species to another wouldn't have much effect since the only parts that differ are the parts the aren't necessary.

Thirdly, "One might suggest, only half facetiously, that the (still to be determined) congruence of their phylogenies does not mean that they have never been transferred but, radically otherwise, that they have always been transferred together."

The idea here is that the logic of the arguments made by Three Domain supporters does not stand up to close scrutiny. It seems superficially reasonable but falls apart when you poke at it. This is the same phenomenon that we witnessed with the Introns-Early Hypothesis. Clearly, Doolittle has been sensitized by his experience in that controversy. (Alternative splicing is another example of bad logic. We'll poke at that one some other time.)

So what is the alternative to the Three Domain Hypothesis? Doolittle says, "We claim that there are no data to contracdict the possibility that every gene we find in any genome today has experienced at least one between-species LGT in the 3-4 billion years since life began." This means there is no single tree of life and eukaryotes and not closer to archaebacteria than to bacteria. The way to represent early evolution is as a complex Web of Life.

Perhaps there is a plurality (most favored) pattern, or one tracked by several genes that we consider important, but this has yet to be proven. In any case, there is no compelling reason why this plurality pattern needs to correspond, by any simple mapping, to the tree of speciations and cell divisions. We cannot infer a unique tree of organisms from the pattern of relationships among genomes without making further assumptions about evolutionary processes that are just that: still-unproven assumptions. We have, for several decades, thought that our job was to uncover the structure of a Tree of Life, whose reality we did not question. But really, what we have been doing is testing Darwin's hypothesis that a tree is the appropriate representation of life's history, back to the beginning. Like any hypothesis, it could be false.

The figure below is taken from Doolittle's Scientific American article "Uprooting the Tree of Life" (February 2000). © Scientific American

---

Microbobial Phylogeny and Evolution: Concepts and Controversies Jan Sapp, ed., Oxford University Press, Oxford UK (2005)

Jan Sapp The Bacterium’s Place in Nature

Norman Pace The Large-Scale Structure of the Tree of Life.

Woflgang Ludwig and Karl-Heinz Schleifer The Molecular Phylogeny of Bacteria Based on Conserved Genes.

Carl Woese Evolving Biological Organization.

W. Ford Doolittle If the Tree of Life Fell, Would it Make a Sound?.

William Martin Woe Is the Tree of Life.

Radhey Gupta Molecular Sequences and the Early History of Life.

C. G. Kurland Paradigm Lost.

---