How the Folding of Proteins Is a Huge Problem for Darwinian Evolution (with Dr. Douglas Axe)

Protein Folding and the Search for New Genetic Information

It's fascinating stuff. Already, you're a little over my head, but I'm trying to keep up. But it's very interesting. So, you're doing your work, and is this where your work has been? So, on proteins?

Yes. Not on this process of biological synthesis of proteins, but on the question of how much information is contained in a gene. In other words, how hard is it to get a new gene that would make a new one of these, is really where I've been focusing.

Right, exactly. And, I mean, you not only were able to show that the formation of the protein is extremely rare, but you also showed that they're not evolving anymore?

Well, here's how we did it. So, the first project was looking at how much can you mess up in a gene and still get the protein. We didn't talk about folding here, but that synthesis process that we talked about that uses the genetic code just really gives you a string of amino acids.

It turns out that you have to have a very special sequence of amino acids in order for that string to collapse and form a stable three-dimensional structure, and that's called the protein folding process.

So, in cells, almost all of the proteins—well, all of them, in their functional form, are going to be folded into a specific three-dimensional structure, and we've now cataloged thousands of these different structures, and you can look at the beautiful match between the shape of the thing and what it does.

Like, if it has to bind to another protein, then the two proteins will fit together, and you can see that in their shapes. If it's doing chemistry, then the molecules that it's working on will bind into a groove inside the protein.

So, all of this is done by folding to the right shape and then having the right chemistry within that shape. So, my work has been looking at, for a long time, how hard is it to get the sort of sequence that will do this at all, and I did that by taking one of these folded proteins that does chemistry—it's an enzyme—and shuffling, really, the amino acids within different regions of this, and then putting bacterial cells that have the shuffled versions into a test environment and scoring them, finding out what fraction of those that have been shuffled still work.

So, you can get a fraction by doing that, and by making some careful assumptions about what you're looking at, you can then use those fractions to infer, well, what if the whole thing had been shuffled?

What fraction would work then? And when you do that, you end up with a very, very, very tiny fraction, something like one in a trillion, trillion, trillion, trillion, trillion, trillion. Very, very small.

So small that there really should be no way that these things are stumbled upon by accident.

Mm-hmm. And then you ended up taking your findings and it was published.

That particular finding was published in the Journal of Molecular Biology in 2004.

Protein Folds and the Limits of Mutation and Selection

And how was that?

You also talked about the—you also talked about the idea that it's not—that things aren't evolving anymore. And that came from another set of studies that we've done with colleagues. So, Anne, Gajer, and I look at—you can think of that first result as measuring the difficulty of stumbling upon an entirely new folded form.

It's called a fold.

Yeah.

Because to get that, you have to get the gene that has the sequence right to fold, and that's really what I measured to be one in 10 to the 77th or 74th power. One in a trillion, trillion, trillion, trillion, trillion, trillion.

Another question is, if you look at these natural enzymes, you will find examples where the two—there are two enzymes that are doing different functions. We'll call them A and B. But their structures are so similar that you could arrange them on a computer and align them, and they look very much like—they're very similar.

They're not identical.

Mm-hmm.

But we say they have the same fold, so the overall structure is very similar. And we've been asking whether you can, with mutation and selection, get a transition from A to B function. And a lot of other people have been doing work in this area, but they're usually assuming that this explains how all these A and B enzymes came to be in the first place, whereas we're willing to say, okay, let's not assume that.

Let's just go see if you can get an A to convert into a B. And when we did this, when we picked what seems to be a very favorable case, we find through extensive study that, in fact, you can't get the A that we looked at to convert into B, not in billions of years.

So the evolutionary story that these transitions work, we've called into question, and we say there's actually no reason to believe they do work. Now when we do that, one of the responses—it's really become the main response—has been well, enzymes don't evolve anymore.

There was an older version, so you're looking at A and B, but those are both modern enzymes. There was some older version, a C or an A prime or something, that gave rise to both of those. And that was an evolvable protein, but the modern ones don't evolve anymore.

They've been so perfected by natural selection that you can't expect them to evolve. But when we're scratching our heads saying, okay, so this is now a hypothetical theory about hypothetical proteins.

We were testing initially a hypothetical idea about real things, but now it becomes a hypothetical idea about hypothetical things, which becomes very hard to put that to the test.

Why the “Modern Enzymes Don’t Evolve” Reply Feels Ad Hoc

Yeah. Kind of reminds me of punctuated equilibrium or something, where it seems ad hoc, maybe.

Yeah, it's a term you give to something where you should really step back and say, maybe we don't know what we're talking about.

So, I mean, interacting with people and giving them the benefit of the doubt. Again, people on the other side of the fence. I tend to see this, and you even bring up Neil deGrasse Tyson as an example.

But it's the science. It's the science. Wherever the science leads, that's kind of like the marching beat there. But when you then show something that is quantifiable, I mean, something that you've done in the field, why not accept?

Were there any evolutionists that looked at this and said, wow, this is a problem, and we're going to wrestle with this and digest this finding?

People tend not to say that publicly. So, some of them might have been scratching their heads and saying, okay, this sounds like it's a problem. I interpret it as having been a problem when the response is one that looks kind of desperate.

So, saying that, okay, genes no longer evolve, proteins no longer evolve, but they did once, that sounds desperate to me. So, it sounds as though someone did recognize that there's a problem. At least what has happened, and I take some comfort in this, no one has come back and said, you're wrong.

You messed up. That A to B transition works perfectly well. And I think the reason no one did that is because they knew that if they said that, we would say to them, show us the mutations, because we've now studied this for years.

If you think this works, show us how it works. And neither has anyone said that we're wrong to say that these things are not readily accomplished, because this new position that, in fact, modern enzymes are not evolvable the way the old ones were, is really an acknowledgment that no one is going into the lab and finding that these things work the way we thought they would work 20 years ago.

And now that so many people have studied these, they really don't work.

Darwin's View of Natural Selection as a Constant Process

Right. And then two, originally, is this not, I mean, Darwin's proposal originally, and the way that it was understood originally as it was accepted by the scientific community, I mean, there was a fluidity that is supposed to go on ad infinitum, right?

Sure. So he described natural selection as being something that's constantly scrutinizing life, constantly sifting, constantly looking for the better variant, and constantly replacing the worst one with the best one.

So he viewed it as something that was always at work. And I think even your sort of traditional orthodox Darwinist to this day, Richard Dawkins, would certainly believe that it's constantly at work. He would, I don't think he would subscribe to the idea that modern genes no longer evolve.

I think he would find that to be strange, and a number of sort of traditional orthodox Darwinists would as well. But the people are actually dealing with the data, having to come to terms with the fact that no one can actually show how this evolution works.

And so we have to come up with a way of saying that we got here by a process that did work, but the way we got here has made it so it no longer does work.