Evolution Icon Evolution
Intelligent Design Icon Intelligent Design

In Explaining Proteins (and Life), Here’s What Matters Most

Claude_Monet._Haystack._End_of_the_Summer._Morning._1891._Oil_on_canvas._Louvre,_Paris,_France.jpg

University of Toronto biochemist Larry Moran and I continue to disagree about what constitutes an assumption versus a fact, what a straw man argument is, and what matters most in the explanation of biological diversity (see hereherehere, and here). The conversation began with the publication of our latest paper, which Larry has avoided discussing. In spite of differences, though, there are some things we agree on. 

Of Straw Men

Larry Moran accuses me of having constructed a straw man argument about the evolution of enzymes. In fact my argument was derived from a widely cited paper that I quoted in my post.

I didn’t misrepresent their argument. Rather than restate it here, I urge you to go back to the original post and see for yourself whether the argument’s made of straw. 

At issue is whether or not evolution is true. By that Larry Moran means the origin of all living things, not just some promiscuous enzymes. He says yes, evolution is true, definitely. I say that’s the point under discussion. It’s backwards to state something as fact and then give examples as to how it might have happened, then end with, "It’s reasonable to conclude" that this is the way it happened. I understand he sees his examples as plausible history and evidence of evolution. But his whole argument depends on the bald statement at the beginning that the thing being argued about is true. 

Agreement?

So what do we agree on? You may be surprised (then again, maybe not for those who follow this closely) that we agree on a narrowly defined definition of the word evolution, and we agree that this form of evolution really does occur.

Larry Moran lists as incontrovertible evidence for evolution small changes observable in real time. Perhaps to his surprise, I agree that small changes over time occur, things like antibiotic resistance, or beak size, or stickleback armor, or wing coloration. Those are the result of selection operating on variation in the population, perhaps coupled with a new mutation or two, and environmental effects. Typically these changes are reversible as soon as selection pressure is removed. Yes, of course allele frequencies change, as they do in the examples I just gave.

So much for our agreement. What about the big changes? Can they be arrived at by incremental change? That I do not accept. And that meaning of evolution is what should be under discussion. 

Enzyme Evolution (Small Scale)

So now, let’s address enzyme evolution and the divergence of enzymes to produce related families and superfamilies. Larry Moran says that modern enzymes evolved by specializing from a promiscuous ancestor. As evidence, he says modern enzymes can sometimes catalyze reactions with several substrates (the chemicals they bind to and change), and that it is possible to shift these enzymes to favor one substrate over another. He gives several examples or provides links to them.

Here’s another place where he and I agree. Promiscuous enzymes can be shifted with just a few mutations to a new reaction specificity, provided the capacity for the reaction already exists in the starting enzyme, and each step is small and selectable. They can evolve easily, because they can already carry out the reaction in question. Larry Moran’s description of the process is actually quite good, despite the digs he takes at us. 

It strikes me that Larry Moran would know we agree with him on these points if he had read our papers.

On Promiscuous Ancestral Enzymes

What about reconstruction of ancestral proteins? Do they in fact demonstrate that ancestral enzymes were promiscuous in the ancient past and can therefore account for the broad diversity of enzymes we see today?

In a previous post discussing enzyme promiscuity, Larry Moran makes the statement (emphasis added): 

The difference is that in this scheme there never was a time when an enzyme with one specificity was transformed into an enzyme with a different specificity. The common ancestor could catalyze both reactions. This has implications for protein engineering and for those studies that try to mimic the evolutionary in the laboratory. You can’t transform one enzyme into another without going through an intermediate with relaxed specificity.

Yes indeed. Protein engineers know that in general if you mutate an enzyme, most often the enzyme will do its job more poorly. If you widen the cleft of the active site, it can accommodate larger substrates, especially if they are related in structure. You end up with an enzyme that can do several things poorly. That’s called being promiscuous.
The unspoken assumption in Larry Moran’s paragraph is that ancestral enzymes were all promiscuous. My question is, how do you get new things if you can never transform one specificity to a new one?

The current trend toward ancestral reconstruction studies was started by Joseph Thornton’s work. He used ligand-binding proteins for his study, which are easier to tinker with (his words) than enzymes. Here is an outline of the protocol used in one paper. I apologize if it gets rather technical. The thing to note is the number of interventions and choices that were made to come up with these ancestral sequences. 

His lab used sequences from related species to produce an ancestral tree, choosing which portions of the protein sequence to align, and throwing away those that did not conform or that had too many long branches on the tree. They generated multiple trees, then chose from among them the trees with the fewest discrepancies (gene gains or losses); in the remaining sequences they eliminated insertions and deletions that were present only in some of the sequences. They then did extensive checking of their modeled sequences against known structures to make sure they had the same folded protein structure, throwing out any with unusual conformations and keeping only those that fell within a set of parameters they established. They then compared these sequences to known protein sequences, and assigned them phylogenetic identities based on their similarities. They then sought to identify what ligands the proteins bound and with what affinity. When a result was something they expected (a promiscuous binding protein able to bind both substrates), they took that as confirmation that they generated the ancestral sequence.

I was not able to find a comparison of the starting sequence to the final sequence of each of these proteins, so I cannot say where or how many mutations were introduced. One table indicated on the order of 30 percent identity, which is low. I’d be interested in more information about where the differences lie.

The resulting proteins did bind ligands and show some activation activity but very low compared to the starting protein. This was taken as a sign of promiscuity, and of success. Were they the true ancestral form? I don’t know.

To my mind this paper is the best of the ancestral studies. Thornton’s lab has done much interesting work since then on structure/function relationships in his proteins. He’s found five specific amino acids in his ancestral protein that are needed to gain the modern function, yet they destabilize the ancestral fold. It’s a catch-22 situation. As you move toward the modern enzyme you destabilize the protein fold, meaning it ceases to function.

�To quote the abstract:

Unless these ratchet-like epistatic substitutions are restored to their ancestral states, reversing the key function-switching mutations yields a non-functional protein. Reversing the restrictive substitutions first, however, does nothing to enhance the ancestral function. Our findings indicate that even if selection for the ancestral function were imposed, direct reversal would be extremely unlikely, suggesting an important role for historical contingency in protein evolution.

To put it briefly, that’s a severe constraint on the evolutionary path.

Some have said that because reported "ancestral" proteins form a stable fold, that indicates they are genuine ancestors. Sometimes "ancestral" proteins don’t fold, however. I’m thinking of the study with one TIM beta/alpha barrel enzyme called HisF. They divided the enzyme into four structural parts because that is how they thought it had evolved. The researchers then created the ancestral tree by, as they put it, throwing out any sequences that didn’t conform to the canonical tree. They calculated what sequence was ancestral to each of the four parts, and made the proteins in E. coli. Two of the four parts were insoluble or degraded. They went on to assemble the pieces that worked into a barrel-like structure by adding cross-linking disulfide bonds. Unfortunately there was no mention of whether or not it still functioned as an enzyme. What does that prove about evolution? It says more about the ingenuity of humankind than about ancestral proteins.

What does it all mean? If you make an "ancestral form" and it is promiscuous, that doesn’t mean all ancestral proteins were promiscuous, or that you have the right ancestral sequence. In the end, it doesn’t matter much anyway, because the biggest problem remains unsolved. 

The Big Problem

Here’s the big problem — the arrival of novelty.

Novelty or innovation means the appearance of something not already present. It’s the opposite of promiscuity. So a way to create novelty is absolutely essential to explain modern cells, as I will demonstrate.

The stripped-down first minimal cell would have needed on the order of 300 genes to survive. The 300 essential genes of E. coli make a good approximation of what genes the first cell must have contained. Among E. coli‘s 300 essential genes, there are 228 unique domains (units of protein structure). That’s quite a lot actually.

A minimal cell with 300 genes already has solved most of the major problems facing new life: how to replicate DNA and divide, how to make membranes and many cofactors for metabolism, how to transcribe DNA into RNA and translate RNA to proteins, how to make ribosomes, etc. But it lacks the enzymes to make things like proteins and nucleotides, and other components of metabolism.

These things are present in the entire complement of modern E. coli‘s annotated genes, however; the cell has 1400 enzymes now, and 606 unique protein domains. That means there are almost 400 new domains in the modern cell that were absent in the ancient cell. They couldn’t have come from promiscuous enzymes, because promiscuous enzymes when they diverge functionally do not diverge structurally. They still keep the same domain structure. In fact, you can’t get a new stable domain by tweaking something already existing. So where did these new domains come from? It wasn’t by horizontal gene transfer. New domains are new topological structures. The first cells likely would have shared most of their genes in common, having to solve the same basic problems, and coming from the same common pool of resources. Even among modern proteins, there are unique domains found only in one organism, or one family of organisms. There will always be novelty to explain.

Here’s the heart of the matter. Promiscuity cannot solve the problem of novelty. Mutation, natural selection, and drift cannot drive the creation of novelty of all those new protein folds. That’s what Doug Axe and I have been testing all along, from Doug Axe’s 2004 paper to this most recent one. Based on our experiments, the problem of how innovation originates remains unsolved.

Image: Claude Monet [Public domain], via Wikimedia Commons.

Ann Gauger

Senior Fellow, Center for Science and Culture
Dr. Ann Gauger is Director of Science Communication and a Senior Fellow at the Discovery Institute Center for Science and Culture, and Senior Research Scientist at the Biologic Institute in Seattle, Washington. She received her Bachelor's degree from MIT and her Ph.D. from the University of Washington Department of Zoology. She held a postdoctoral fellowship at Harvard University, where her work was on the molecular motor kinesin.

Share

Tags

Science