Protein folding can be stable, or it can be a whirlwind of instability—but natural selection can reap function from both.
This is a companion discussion topic for the original entry at https://biologos.org/blogs/dennis-venema-letters-to-the-duchess/reaping-the-whirlwind-protein-function-without-stable-structure
Great article, Dennis! Just one little thing: it describes Axe’s calculated probability as 1 in 1077, rather than 1 in 10 77 .
Best,
Chris
2 Likes
That’s a perennial formatting problem. I’ll flag @BradKramer to alert him to the need to fix it. Thanks!
Dennis, I’ve just gotten your new book. Haven’t gotten around to cracking it open yet, but I must say: whoever designed the cover deserves an award of some sort. Very good looking book. 
2 Likes
My colleague down the hall is one of the leaders in the field of IDPs. It was quite a juxtaposition reading Axe’s book while walking by posters from his group’s work every day.
4 Likes
Great post! IDPs are really interesting. My favorite new role for them is to allow tardigrades to live forever via desiccation. My friends at Molecular Cell published the paper and it made the New York Times.
Tardigrades Use Intrinsically Disordered Proteins to Survive Desiccation (journal article, contact me for PDF if interested)
How One Of The World’s Toughest Creatures Can Bring Itself Back To Life (story on Morning Edition)
How a Water Bear Survives, Even When It’s Dry (NY Times piece)
4 Likes
Are you judging a book by its cover?
2 Likes
Tardigrades are my ten-year old son’s favorite animal, whenever anyone asks. So far only a couple people have known what he is talking about. We’ll have to listen to the Morning Edition story together. Thanks for the link.
1 Like
What can I say, I like books with good looking covers. Design has sort of been a side interest of mine. A well-designed book cover can go a long way. 
2 Likes
Intelligently designed, you might say.
2 Likes
Oops! @DennisVenema even left me a note about this, and I still forgot. -100 editor points for me. Should be fixed now.
2 Likes
@DennisVenema ,
Great series. The ability of proteins to have function without stable folds was new to me, and definitely opens the aperture wider for evolutionary steps.
That said, I’d like to revisit the issue of nylonase. As you have noted in earlier posts, and referred to again here, Ohno in 1984 claimed that nylonase originated from a frameshift mutation which gave rise to a radically new/different protein, with a new fold that happened to be somewhat active for nylon decomposition; this activity was subsequently enhanced by a few additional, more routine (presumably stepwise) mutations.
However, there seem to be a number of sources which challenge the notion that a frameshift was involved in making the initial nylonase. This challenge seems to stem largely from the 2005 Negoro paper [Journal of Molecular Biology Volume 370, Issue 1, 29 June 2007, Pages 142–156 ; X-ray Crystallographic Analysis of 6-Aminohexanoate-Dimer Hydrolase - Journal of Biological Chemistry
] which I know you are aware of. Granted that this paper focuses on the last two mutations to obtain today’s highly effective nylonase. But it seems to claim that the starting point for this mutations was not a novel frameshifted protein, but rather an ordinary, preexisting esterase. Perhaps because the authors were Japanese, they were too polite to come out and say “Ohno was wrong” in print. However, they did say as much to an interviewer from New Scientist, who wrote [ Five classic examples of gene evolution | New Scientist ]:
“ Nylon was first made in 1935. Just 40 years later, in 1975, a bacterium was discovered that is able to digest and live off not nylon itself, but waste chemicals from its manufacture – chemicals that had not existed before nylon production began.
It was later shown this bacterium, now known as Arthrobacter KI72, has evolved several types of enzymes capable of utilising these waste products. One type, 6-aminohexanoic acid hydrolase, encoded by genes called nylBs, has become known popularly as “nylonase”.
As a dramatic example of evolution in action, nylonase has attracted a lot of attention over the years. But there has also been a great deal of confusion about how it evolved.
In 1984, geneticist Susumu Ohno suggested that one way in which new genes could evolve is through a “frameshift” mutation – one that alters the way in which the genetic code is read and thus completely alters the amino acid sequence of a protein. And nylonase evolved this way, he claimed.
Then in 1992, another team claimed that nylB genes are unique and had evolved by a rather complicated and special mechanism.
They are both wrong, says Seiji Negoro of the University of Hyogo, Japan, whose team has published many studies on the structure and evolution of nylon-related enzymes. “I believe that the above two hypotheses can be excluded,” he told New Scientist.
His team’s study of the protein structure show that nylonase is very similar to a common type of enzyme that breaks down beta-lactamases – natural antibiotics produced by many organisms. Just two amino-acid changes – two mutations, in other words – are required to change the beta-lactamase binding site to one capable of binding the nylon by-product.
However, while Ohno was wrong about nylonase, he was right about frameshift mutations being one way in which genes evolve. Hundreds of examples have now been discovered in humans alone.”
The Wikipedia article “Frameshift Mutations” likewise claims that the frameshift concept is likely wrong:
“Frameshift mutations have been proposed as a source of biological novelty, as with the alleged creation of nylonase, however, this interpretation is controversial. A study by Negoro et al (2005) found that a frameshift mutation was unlikely to have been the cause and that rather a two amino acid substitution in the active site of an ancestral esterase resulted in nylonase.”
Your jousting colleagues at the Discovery Institute have picked up on this and used it to specifically try to rebut your assertions. In March, David Klinghoffer authored, “Nylon and Nylonase: Ann Gauger Disentangles an Evolutionary Icon” [ [https://www.evolutionnews.org/2017/03/nylon-and-nylonase-disentangling-an-evolutionary-icon/]] , writing:
“… When Japanese scientists in 1975 discovered such a bacterium at work, Darwinists later brandished this. They said it showed evolving new proteins is a breeze.
The nylon-degrading enzyme nylonase arose in the course of 40 years. That sounds like a score for evolution, right? Indeed, our theistic evolutionary friends over at BioLogos were in a triumphant mood on the subject a little while back on their Open Forum (“Biological Information and Intelligent Design: evolving new protein folds“). […here they quote from your article where you state that “the fact that we can observe new functions coming into being” shows that the Axe/Meyer estimates of the improbability of the are grossly over-inflated…]
…In a brief window of time, some bacteria developed the ability to consume nylon. Ergo, no need for intelligent design in biology?
Not so fast, explains Discovery Institute biologist Ann Gauger. In an ID the Future conversation, she talked with Sarah Chaffee. Conclusion: ‘Nylonase was a pre-existing enzyme, had a pre-existing activity. It was easy to convert it to the ability to degrade nylon [by a] step-wise path. Therefore, there’s no reason to think that the enzyme is a newly derived enzyme from a frame shift. We don’t need that explanation.’ In short, as an icon of evolution, nylonase has no legs. “
The Klinghoffer article concludes with links to a podcast consisting of an interview with Ann Gauger. Her conclusion was stated in the previous paragraph. She dismisses the notion of a frameshift mutation as merely a “hypothesis”, for which there is no hard evidence. Here is some of my transcription of part of her talk, which may not be 100% accurate word for word:
“What Ohno noticed about nylonase was that the DNA had two reading frames forward that did not have stop codons in them. So he speculated that there was a pre-existing gene and it inserted a start codon into a different frame and then you got the new reading frame…
So,he was suggesting that you got a frameshift mutation that completely changed the coding sequence of the DNA to a completely new protein with a function. So he was suggesting it was not hard to get a coding sequence with a function. Just shift the frame and you’ll have a protein that folds and it can carry out some sort of activity.
Now you can see how that would go directly against our suggestion that it’s hard to get new information that’s functional, that it’s really really hard to get a new functional fold, a new functional enzyme for example…
Question from the interviewer: “Was Ohno correct?”
It,was an interesting hypothesis he put forth in his paper …but it was a hypothesis only , and it’s been taken as fact by many evolutionary biologists, at least the ones that write on the web. But it turns that there’s more to the story which doesn’t get publicized, which is significant.
The Japanese workers who have been working with nyl B continued their study, they purified the protein , crystallized it, determined its structure, its three dimensional fold, and discovered that the protein encoded by nylon B had a fold structure similar to other known enzymes, carboxyl esterases, they are called. Carboxyl esterases degrade a particular bond that’s similar in structure to the nylon bond that nylon ace degrades. So they thought, OK, nylonase looks like a carboxyl esterase, let’s see if it has carboxyl esterase activity, and lo and behold it did. This is suggesting that nylonase was a pre-existing enzyme that had a different functional activity , that was involved in degrading carboxyl esters, and it already had that function, so then they looked if they could say anything about how it has evolved the ability to degrade nylon.
… They did some mapping work and identified that to convert from carboxyl esterase to the ability to degrade nylon took only two amino acid changes. One was sufficient to convert to nylonase, and the second improved the nylonase activity. … [ordinary, microevolutionary] Darwinian processes are capable of doing that…
Sorry this has gotten kind of long. But I think it is important to thrash this issue out. If the starting esterase (which just happened to have a fold resembling other esterases) really sprang into existence de novo from a frameshift in the space of 40 years, of course that is a nice example of the power of evolutionary processes. On the other hand, if the starting esterase was just an ordinary enzyme that happened to be part of the bacterium prior to the advent of nylon, that should be made clear, and we can move on. Not being a biologist, I’m not sure how this gets settled. Thanks…
I’m a bit late in this dicussion, but I just now read this as part of taking interest in the stochastic model offered by Doug Axe in his 2004 Article in JMB and the arguments offered by the ID community against evolution. I noticed that there are some rebuttals similar to yours published in blogs, etc. My background is systems engineering where I’ve worked on a variety of complex systems, both designing and maintain them, e.g, science payloads for the space shuttle. From that perspective I find the arguments against ID seem to not reflect an understanding of one of the most fundamental tenets of systems engineering, viz., that changes to a complex system of systems must be coordinated. Thinking about the evolutionary process of random change(s) to the code sequence in DNA, most seem to think that a small change to the DNA will result in a slight change to a protein that is specified by it and perhaps also a small change to the way or number of ways a protein may fold. Your argument supports the notion that these small changes may not be destructive, but could be the foundation for evolution. The point that I would like to make for your consideration is that for each small random change in the DNA, a corresponding change must also occur in all the other sub-systems that depend on the one that experienced the random change if they are to remain functional at the higher “system of systems” level. You simply cannot change a single protein even if it only results in a minor fold change or makes it unstable so that multiple folds are possible. The effect of the change cannot be realized unless all the other proteins change in very specific ways that enable the required coordination to continue in the cell that depend on that protein’s function in all of its states. (This last comment may explain what you observe as IDPs. Has anyone investigated that possibility that a protein may have multiple functions depending on it “machine state.” If you’re seeing a single protein in multiple folded configurations, can it be that those are actually necessary and designed into it? The proteins may, through command and control (C&C) signaling, change states to perform a different function. Our inability to detect the C&C signaling and trace it across the cell to see its full impact may be the reason you’re willing to conclude that it’s an evolutionary mechanism. Instincts from my SE background tell me that random changes that are uncoordinated across all the cell’s sub-systems (other proteins or collection of proteins) are absolutely (100%) destructive. With perhaps as many as a thousand different sub-systems in the cell that would need to change in coordination with the mutated change in a single protein, the probability of that occurring is likely very small. (I haven’t tried to create a stochastic model for this SE effect but I suspect it would show that Axe’s estimation is grossly underestimated.)
Your thoughts?
We know this to be false, because we know a fair amount about various sources of robustness in molecular biology (and at other levels). But even if you didn’t know about robustness, you should know that there is substantial variation in protein structure in living organisms right now. The notion that one little change requires some epic “coordination” to avoid catastrophe… is falsified every time an animal (including a human) produces live offspring.
Your critique misses the mark based on some pretty basic biology. It is always so disappointing to see comments like this, which not only display ignorance but which come with the assumption that evolutionary biologists have ignored obvious “problems” with the theory.
Welcome to the forum, Raymond!
@Chris_Falter @DennisVenema, @jammycakes @Argon @T_aquaticus or any other science types have answers to the question above?
Welcome to the forum, Raymond. Stephen tends to be blunt, which is part of his charm, but as a borderline biologist, I think he is referring to the fact that in biology it is not an all or nothing thing with different configurations of molecules, but varies. An easy to see real life example is the use of a variety of different insulins including porcine in the treatment of diabetes. While they have different properties and rates of metabolism, they all lower blood sugar and are useful in treatment. The same can be seen in antibiotic resistance where incremental changes in a resistance mechanism can develop with repeated mutations.
It is interesting that engineers tend to be disproportionately attracted to ID arguments, probably as a result of training reinforcing an underlying tendency to look for order, but that is another topic.
1 Like
Stephen is being blunt, yes, but he’s correct. If the critique offered was valid, then normal variation in populations would create huge problems. It doesn’t. Proteins are flexible, and changes in their structure are often tolerated without issue. The rigid “lock and key” analogy for protein binding is misleading.
Biological systems are unlike human-designed systems in many ways. Because of this, your instincts are misleading in this case.
2 Likes
Well, I can be a bit harsh too, and I am sorry about that. @Raymond_Isbell, please don’t let my gruffness discourage you from discussing proteins or evolution or (my current preference) chocolate with me.
2 Likes
I think I’ll leave it to the biologists. I’m more a physical sciences guy (which is why I have more to say about the age of the earth than about evolution) and my general understanding of biology is pretty rudimentary.
I don’t think that’s surprising really. After all, engineering, by definition, is designing things – and very often these things require one to exercise a lot of one’s own intelligence in order to do so. It’s difficult to have a problem with Intelligent Design when “intelligent design” could just as well be your own job description.
2 Likes
Speaking for myself, I sometimes think my field in primary care medicine is similar.
1 Like