Behe's Two-Binding-Sites Rule
by BilboIn his book, The Edge of Evolution; the Search for the Limits of Darwinism, Michael Behe tries to find where exactly the limit to Darwinian evolution is. In a previous thread, Behe's Test, Take 2, it was admitted that if it takes more than two mutations (with the question of whether this includes neutral mutations being brought up by not discussed at length) before a selective advantage is bestowed, then Darwinian evolution probably wouldn't happen. The question is whether or not more than two mutations have ever been needed for evolution to occur. Behe would say, "Yes." And in Chapter 7, "The Two-Binding-Sites Rule," he presents his argument. First there is a long discussion on the nature of "shape space," and then Behe gets to his argument, beginning on page 133:
Consider a hypothetical case where it would give an organism some advantage if a particular two of its proteins, which had been working separately, bound specifically to each other. Perhaps the two-protein complex would be able to perform some new task, or do an old task much better. The lesson from shape space is that, in order for the one to bind the other, we should expect to have to search through tens of millions of different mutant sequences before luckily happening upon one that would specifically stick with even modest strength, which would allow the two to spend even half of their time together. (This is likely the minimum necessary strength, enough to have a noticeable biological effect.) [11] Since the mutation rate is so low — about one mutation at a particular site in a hundred million births — we would expect to have to slog through an enormous number of organisms before striking on that lucky one.
Let's make a rough calculation for the average number of organisms we would have to slog through to find a new protein-protein binding site. As I said, shape space tells us that about one in ten to a hundred million coherent protein-binding sites must be sifted before finding one that binds specifically and firmly to a given target. The simplest way to alter a protein is by point mutation, where one amino acid is substituted for another at a position in a protein. There are twenty different kinds of amino acids found in proteins. That means that if just five or six positions changed to the right residues — the ones that would allow the two proteins to bind — that would be an event of approximately the right frequency, since twenty multiplied by itself five or six time (20<5> or 20<6>) is about three million or sixty million, respectively — relatively close to the ten to a hundred million different sites we need.
So one way to get a new binding site would be to change just five or six amino acids in a coherent patch in the right way.[12] This very rough estimation fits nicely with studies that have been done on protein structure.[13] Five or six amino acids may not sound like very much at first, since proteins are often made of hundreds of amino acids. But five or six amino acid substitutions means that reaching the goal requires five or six coherent mutational steps — just to get two proteins to bind to each other. As we saw in the last chapter, even one missing step makes the job much much tougher for Darwin than when steps are continuous. If multiple steps are missing, the job becomes exponentially more difficult.
Let's consider one further wrinkle. Most amino acid changes in proteins diminish a protein's function. But about one-third of possible amino acid changes are like switching a k for a c in "cat" or "candy"; they can be accomodated without too much trouble.[14] Such "neutral" changes can occur during evolution and spread around a population by chance. So let's suppose that of the five or six changes that have to happen to a protein to make a new binding site, a third of them are neutral. They could occur before the other key mutations, as a separate step, without harm. Although finding the right neutral changes would itself be an improbable step, we'll again err on the conservative side and discount the average number of neutral mutations from the average number of total necessary changes. That leaves three or four amino acid changes that might cause trouble if they occur singly. For the Darwinian step in question, they must occur together. Three or four simultaneous amino acid mutations is like skipping two or three steps on an evolutionary staircase.
Although two or three missing steps doesn't sound like much, that's one or two more Darwinian jumps than were required to get chloroquine resistance in malaria. In Chapter 3 I dubbed that level a "CCC," a "chloroquine-complexity cluster," and showed that its odds were 1 in 10<20> births. In other words (keeping in mind the roughness of the calculation):
Generating a single new cellular protein-protein binding site is of the
same order of difficulty or worse than the development of chloroquine
resistance in the malarial parasite.Now suppose that, in order to acquire some new, useful property, not just one but two new protein-binding sites had to develop. A CCC requires, on average, 10<20>, a hundred billion billion, organisms — more than the number of animals that has ever existed on earth. So if other things were equal, the likelihood of getting two new binding sites would be what we called in Chapter 3 "double CCC" — the square of a CCC, or one in ten to the fortieth power. Since that's more cells than likely have ever existed on earth, such an event would not be expected to have happened by Darwinian processes in the history of the world. Admittedly, statistics are all about averages, so some freak event like this might happen — it's not ruled out by force of logic. But it is not biologically reasonable to expect it, or less likely events that occurred in the common descent of life on earth. In short, complexes of just three or more different proteins are beyond the edge of evolution. They are lost in shape space.
And the great majority of proteins in the cell work in complexes of six or more. Far beyond that edge.
(pp.133-135)
I'm not sure if Behe's argument is a good one or not. He discusses in an endnote the option of other mutations besides point mutations, which I can also copy if needed. He doesn't discuss the difference between simultaneous and consecutive mutations. And the power of his argument seems to come down to his last two sentences, which is rather reminiscent of his argument for irreducible complexity. I think if people granted that the first living cells needed the great majority of proteins to work in complexes of six or more, then his argument would be difficult to refute. But how many of you critics out there are willing to grant that? Probably none of you, is my guess. But I'm still curious to hear your comments.
I'm adding Behe's endnote for [11]:
Most proteins are present in the cell at well below millimolar concentrations, so in order for two proteins to spend the majority of their time bound to each other, micromolar dissociation constants would be required to form even a "weak, transient" complex (Nooren, I.M., and Thornton, J.M. 2003. Structural characterisation and functional significance of transient protein-protein interactions. J. Mol. Biol. 325:991-1018). Dissociation constants on the order of micromolar seem to be required to detect interactions in yeast two-hybrid assays. (Golemis, E.A. and Serebriiskii, I. 1996. Identification of protein-protein interactions. In Coligan, J.E., ed. Current protocols in protein science. Brooklyn, N.Y.: John Wiley & Sons, Inc; Estojak, J., Brent, R., and Golemis, E.A. 1995. Correlation of two-hybrid affinity data with in vitro measurements. Mol. Cell Biol. 15:5820-29).
August 28th, 2008 at 1:23 pm
To be clear from the outset, Behe is talking specifically about cellular protein-protein binding sites, so critics can discard the viral binding sites refutation.
Comment by chunkdz — August 28, 2008 @ 1:23 pm
August 28th, 2008 at 3:40 pm
chunkdz — nice attempted pre-emptive goal-post rolling, but unfortunately for you Behe made HIV one of his major examples, so virus binding sites are clearly "in".
Bilbo —
Basically:
1. Biologists would not accept that anything beyond 2 simultaneous mutations is so improbable as to be impossible, there is a paper somewhere that estimates that in the global bacteria population, multiple simultaneous mutation combinations are happening regularly. Oh yes:
Whitman et al. (1998) Prokaryotes: the unseen majority. PNAS 95(12):6578-83.
==============
Genes that are widely distributed in prokaryotes have a tremendous opportunity for mutational change, and the evolution of conserved genes must be otherwise greatly constrained. Assuming a prokaryotic mutation rate of 4 Ć 10ā7 mutations per gene per DNA replication (86, 87), four simultaneous mutations in every gene shared by the populations of marine heterotrophs (in the upper 200 m), marine autotrophs, soil prokaryotes, or prokaryotes in domestic animals would be expected to occur once every 0.4, 0.5, 3.4, or 170 hr, respectively. Similarly, five simultaneous mutations in every gene shared by all four populations would be expected to occur every 60 yr. The capacity for a large number of simultaneous mutations distinguishes prokaryotic from eukaryotic evolution and should be explicitly considered in methods of phylogenetic analyses.
==============
2. And it is a fallacy to assume that point mutations are the only kinds of mutations, particularly with binding sites there are several other known mechanisms, e.g. domain swapping, duplication/transposition of binding domains, etc.
3. Nevertheless there is no particular evidence that single protein-protein binding sites are very hard to evolve, binding is a quantitative property and it is known that single point mutations can move the binding affinity up or down a few orders of magnitude. And we have numerous examples of binding sites evolving in lab experiments, both with immune system & non-immune system protein systems.
4. The fact that multiple residues *contribute* to *current* binding in a particular system does *not* mean that *all* of those residues used to be absent and were *all* placed there by mutation and selection. The way the evolution of binding sites occurs is more like: (1) all proteins are sticking to each other all the time to varying degrees, then (2) selection favors a particular mutant that strengthens a particular interaction — and even at this stage the single new substitution already cooperates with several other amino acids to produce binding. That first selected mutation has *already* selected a set of nearby amino acids that were preexisting and happened to help produce binding, that is why the new mutation worked in the first place, then (3) further rounds of mutation/selection refine & strengthen the binding.
5. A further fallacy is to assume that there is only one sequence configuration to produce a binding site. Basically you are talking about bumpy/sticky surfaces, there are many, many different ways to get 2 things to stick together.
6. Furthermore, there is no particular reason to think that it would have been a common requirement for *two* or more protein-protein binding sites to evolve *at once*. Behe's argument for this is nothing more than his old IC argument & so suffers from all of the same problems.
For some references & a bunch of counterexamples, see my review:
http://www.pandasthumb.org/arc...
So basically Behe's argument fails at every possible level, it's quite impressive really.
Comment by Nick Matzke — August 28, 2008 @ 3:40 pm
August 28th, 2008 at 4:03 pm
Matzke,
From Edge of Evolution, pg 148
Leave the "gotcha politics" to McCain and Obama, Nick.
As for the rest of your post, I lost count of the strawmen by the third paragraph. It would be helpful if you could actually quote "Edge" verbatim rather than just giving us your slant on it.
Comment by chunkdz — August 28, 2008 @ 4:03 pm
August 28th, 2008 at 4:45 pm
Bilbo:
Why on earth would you think that the first living cells needed that?
I think the best argument against Behe's claims is that binding affinity between proteins is not an all-or-nothing thing but rather a quantitative character that lies on a "continuum" between strong negative affinity and strong positive affinity. A single mutation can increase affinity and confer some selective advantage. Did Behe provide any evidence to show otherwise? I have his book somewhere but I never finished it and am too lazy to look it up right now.
But let's suppose for a moment that Behe is right. Are we to believe that the vast majority of protein-protein interactions that ever evolved in millions of species required divine intervention to supply the ultra-unlikely combination of mutations?
Comment by Raevmo — August 28, 2008 @ 4:45 pm
August 28th, 2008 at 5:30 pm
This "rule" is just a re-wording of the same problem we've been discussing in the previous thread. Behe finds a structure of a certain complexity. He determines that its origin is stupendously improbable, because he assumes that it could not have arisen cumulatively. Same old same old.
My impression of his arguments is that they involve a calculated attempt to confuse. He provides lots of interesting information about a system ā it's complex, it's cool, it's specific, and so on. Then he says the probability of its occurrence is 10 to the zillion, similar to the probability of a "CCC", so — SHAZAM — this thing couldn't have come about naturally. In between the two parts of the magic show, glossed as quickly and as casually as possible, lies the only aspect of the spectacle that matters at all: the question of whether and how the system could have arisen cumulatively.
Folks, that's all that matters, and you can't learn anything by pretending to look hard at examples where an evolutionary change didn't occur. If you want to understand the evolution of something as interesting and specific as a binding site between two proteins, and you already understand common descent, then you have to get to work examining the binding of real proteins and the ways in which the binding sites might have been altered during natural history. You could follow the kind of reasoning that Joe Thornton's group at Oregon followed when studying steroid receptor evolution. You could follow the kind of reasoning that Chris Hittinger and Sean Carroll followed when studying subfunctionalization and adaptive conflict in yeast genes. And first, you might try reading what other scientists have to say about the challenges that Behe thinks he's raised using binding sites.
If you know someone who is interested in doing this kind of analysis, which should be possible at least in principle, contact me. I would gladly collaborate. I have a few ideas regarding places to start. This is a serious offer, which I will repeat on my blog and elsewhere.
If all you really want to do is dismiss all of evolutionary genetics based on a trade book by an uninformed biochemist who doesn't even know John Maynard Smith's surname, then just be careful to avoid reading anything written by a knowledgeable scientist.
Comment by SteveMatheson — August 28, 2008 @ 5:30 pm
August 28th, 2008 at 6:32 pm
Nick (quoting):
Good point. Did I just agree with Nick?
I must be getting soft.
Comment by Bradford — August 28, 2008 @ 6:32 pm
August 28th, 2008 at 6:34 pm
Steve Matheson:
It helps to identify a structure before assessing the likelihood of determining its origin.
Comment by Bradford — August 28, 2008 @ 6:34 pm
August 28th, 2008 at 6:38 pm
Maybe it would have something to do with Bilbo's knowledge of cellular functions. First living cells? They aren't around are they? You can say anything you want about them. Constrained only by your imagination.
Comment by Bradford — August 28, 2008 @ 6:38 pm
August 28th, 2008 at 6:50 pm
Bradford:
Why don't you address the gist of my comment instead of resorting to your pathological hobby of bringing up the first cell in every thread?
Comment by Raevmo — August 28, 2008 @ 6:50 pm
August 28th, 2008 at 7:37 pm
Raevmo:
Pathological!? That's a mighty big word Raevmo. I don't bring it up in every thread. Only when I know you're hanging around.
Comment by Bradford — August 28, 2008 @ 7:37 pm
August 28th, 2008 at 7:40 pm
Raevmo, that was Bilbo, not Bradford. You should limit your posting to before happy hour.
Comment by chunkdz — August 28, 2008 @ 7:40 pm
August 28th, 2008 at 7:47 pm
chunkdz:
Raevmo, let's have a toast to the origin of life. May the event be forever celebrated. Bottoms up. (glug glug glug). Ahhhh. Now you're ready to discuss precursors to DNA polymerase.
Comment by Bradford — August 28, 2008 @ 7:47 pm
August 28th, 2008 at 9:07 pm
Warning:
I am not a biologist and cannot defend Behe's claims. I don't even have a firm grasp on what it means for a protein-protein binding site to evolve.
However, I can refer you to his claims where they seem to contradict how they are being countered here.
Nick Matzke said:
Likewise, it would be a fallacy to claim that anybody ever assumed this.
Behe's discussion does focus on single point mutation and, of course, that is the situation to which his argument is relevant.
The billions of antibodies generated by the few hundred genes makes the immune system example, as presented by Behe, completely different from that of cellular proteins.
Behe doesn't merely assume this. He discusses the variable strength of bonds and the experimental evidence of the shape space one needs to explore to find a suitable site. He makes no assumption about the specificity of the configuration. When it comes to his suggestion that a 5-6 aa sequence is required he compares this to, and possibly derives it from the experimental data regarding the necessary search space to find a moderate strength binding site. It takes a search through ten to one hundred million sites and this would be roughly equivalent to the number of 5-6 aa sequences. page 134
Behe did not provide evidence to suggest that there was not a continuum of binding strength because he didn't make such a claim but rather discussed the search space required in order to find at least a moderate level of binding, defined to be 50%.
He uses the evidence, which he claims on page 132 to have been consistently confirmed experimentally, which says that a given protein must search ten to one hundred million binding sites in order to encounter one to which it will stick with moderate strength.
Given those positions of Behe's continue to interpret and counter him as you will.
Comment by Pez — August 28, 2008 @ 9:07 pm
August 28th, 2008 at 10:05 pm
Bradford:
Exactly. Hence my suggestions for research to address Behe's suggestions. Are you proposing a collaboration? I'm interested in the ideas that the ID community can generate, given its intense interest in the genetics of adaptation and evolution.
Comment by SteveMatheson — August 28, 2008 @ 10:05 pm
August 29th, 2008 at 1:11 am
Off topic — new thread request. Anybody see this article: http://www.livescience.com/hea... It seems to very strongly support the front loading hypothesis.
Comment by bFast — August 29, 2008 @ 1:11 am
August 29th, 2008 at 1:47 am
It fills me with sadness to observe that Bradford and chunkdz's sense of humor is of the same dreadful quality as their reasoning.
Comment by Raevmo — August 29, 2008 @ 1:47 am
August 29th, 2008 at 1:57 am
Hey, where'd my links go?
So, let me try again.
Here's last year's interpretation.
http://www.sciencedaily.com/re...
This was, of course, when the sponge was still considered to have preceded the box jelly.
http://telicthoughts.com/first...
Comment by Pez — August 29, 2008 @ 1:57 am
August 29th, 2008 at 7:19 am
Sup with the devil…
Comment by Alan Fox — August 29, 2008 @ 7:19 am
August 29th, 2008 at 9:34 am
Cheer up then and have that toast with me. What's your favorite brand?
Comment by Bradford — August 29, 2008 @ 9:34 am
August 29th, 2008 at 10:57 am
Raevmo and Nick (and others) have suggested the idea of gradual increase in binding ability between proteins. I'll quote Behe's endnote at length when I have more library time.
Comment by Bilbo — August 29, 2008 @ 10:57 am
August 29th, 2008 at 1:07 pm
Comment by chunkdz — August 29, 2008 @ 1:07 pm
August 30th, 2008 at 5:34 pm
nullasalus,
It's a slightly annoying that I cannot respond to comments directed to me on the appropriate thread, but Bradford leaves me no choice.
Oh, I see that Joy has banned me also from the open thread and deleted my post. Let's try it here. If that doesn't work, then bye bye Telic Thoughts.
You said:
Neither of us knows what Cortunix really says in the classroom. But from parsing his words, I don't think he actually brings up NOMA in the classroom because that would obviously alert students to a possible conflict between the scientific facts he presents and certain religious dogma. So if he doesn't offer it up as truth, he isn't lying. Do you understand that? Rather, he would be doing the responsible thing. Now, he would only be lying if he told the students he believes NOMA to be a valid position while he in fact believes the opposite. That would be unethical. But from reading his post, I don't think he did that. You are jumping to the conclusion that he did, perhaps because you got emotional due to your strong disagreement with his liberal point of view.
Your hypothetical scenario is very creative, but moot because it doesn't mirror at all what Cortunix is doing, in my opinion.
Comment by Raevmo — August 30, 2008 @ 5:34 pm
September 2nd, 2008 at 1:50 pm
Couldn't get to a computer over the weekend. Here's Behe's footnote for [11]:
I'm not sure what millimolar and micromolar concentrations are, nor why this endnote supports Behe's statement that
But if Behe is right (is he?), then it seems to limit the gradual approach to making new proteins, with one amino acid change at a time.
Comment by Bilbo — September 2, 2008 @ 1:50 pm
September 4th, 2008 at 3:26 pm
Even if Behe were right that some arbitrary line he draws is the "minimum necessary strength" for "noticeable biological effect" for "most proteins — which seems extremely unlikely to me (proteins interact in innumerable different ways with different binding strengths, so there are lots of intermediate ways to interact before forming something that is together often enough to be considered a "complex", which itself is a line drawn across a continuum) — it doesn't really matter what is the case for "most proteins", all evolution needs is that very occasionally the preexisting circumstances are good enough in terms of concentration, co-localization, pre-existing weak interaction/binding/accidental sticking, such that a stronger interaction is favorable & favored. Sure, the vast majority of the time not everything happens that would need to happen to produce a new binding site, but this is no more damaging to standard evolutionary theory than saying that most species go extinct or most mutations are harmful.
Here's an example of a common situation where the random-sticking-in-a-loose-soup-of-proteins model is wrong: membrane-embeded proteins. These proteins float around embedded in the membrane, bumping into each other all the time. They already interact to some degree in a default state with the bumping. All it would take to begin to evolve a binding site is some situation where a closer association of 2 membrane proteins caused some improvement, e.g. more efficient transport, or one protein enhances or slows the function of the other in a beneficial way.
Comment by Nick Matzke — September 4, 2008 @ 3:26 pm
September 4th, 2008 at 4:14 pm
Nick,
Say you have two proteins that by lucky chance happen to bump into each other and they weakly adhere. And for the sake of argument let's entirely throw Behe's claim of chemical and geographical protein surface complementarity out the window.
Say that this yields some benefit for the cell.
How do you get it to happen again?
This was post-translational. Will the genes know how to pull off this trick again?
Comment by Doug — September 4, 2008 @ 4:14 pm
September 5th, 2008 at 1:37 pm
Doug, if I understand Nick, he would say that the proteins are already achieving a selective advantage by occasionally interacting. If so, it seems reasonable that RM + NS could enhance this advantage.
But my question is, Is there already a selective advantage in a short duration of interaction (say 10%)? If so, then I think that would strengthen the argument against Behe. I've quoted Behe's endnote on this. I haven't read the referenced articles , so I don't know if they support his statement (about needing to bind 50% of the time to achieve a selective advantage). Is there scientific literature refuting his statement?
Comment by Bilbo — September 5, 2008 @ 1:37 pm
September 5th, 2008 at 4:38 pm
Hi Bilbo,
I was slightly taken aback at what you answered me on Behes test take two. I tought you were taking the micky and trick me into answers that can be ambigious by not so well informed redeas of this bolg. This I find to be wrong when I read
After this, I must admit, I did you an injustice, please acdcept my appologies for it. But scienticts have some perculiiar way of cumunicating that you may not be hep to? I undrstand Behe and his critics but i have my own thoughts. But your admonished controversied and your answers were so far out, I could not react to it.
Pleas accept my appologies if I understood you wrongly in what you tried to convey to me-
alll the best
Karla
Comment by Karla — September 5, 2008 @ 4:38 pm
September 6th, 2008 at 3:09 pm
Hi Karla,
I'm not sure you misunderstood me or not. Since you haven't responded to my questions, there isn't anyway to advance the discussion. If you want to continue the discussion at Behe's Test, Take 2, please do so. It sounds like you know a heck of lot more about this than I do.
Meanwhile, you wrote:
I am curious what those thoughts are, and whether or not they will help us determine if Behe is correct.
Comment by Bilbo — September 6, 2008 @ 3:09 pm