I see. So had he done so, do you hypothesize that would make a difference of 50+ orders of magnitude?
This is, in some sense, the point. We have a highly complex, improbable protein fold performing this function and many others. If evolution can reach this function in a much less complex system, why isn’t that what we observe?
We don’t know the ultimate origin of this protein fold, and hence we can’t really say it did not actually emerge from a much less complex system.
Of course, one possibility is just historical contingency. Among a large number of possible, alternative evolutionary trajectories that were available to the ancestral population that spawned this protein superfamily, this one happened to be the one nearby in sequence space. That from that ancestral position, this outcome was the more likely one.
@BenKissling, before I answer this rather different question, can we agree that the claim that Axe was measuring the frequency in sequence space of folds and not activity is incorrect?
We would expect evolution to produce highly improbable proteins in the same way that nearly all contingent natural processes produce highly improbable outcomes. If you look out the window and see a cloud the exact shape of that cloud is extremely improbable, and yet there it is.
And a lot of them are specified because they form recognizable patterns. Clear example
Of natural causes creating CSI
Why, I believe Dawkins’s weasel program had its origin in the clouds. That’s CSI right there.
can this protein replace the real beta-lactamase?
In the right conditions, yes. In a natural setting there is going to be an outward concentration gradient around the fungal species that is releasing beta-lactams. Even a very limited ability to break down those antibiotics is going to allow the bacteria to outcompete other bacteria and set up a niche in higher concentrations of beta-lactams.
They work while attached to a phage, a type of virus (that infects bacteria) that normally doesn’t even have any antibodies, so yes there’s no reason to think they wouldn’t also work when expressed as any other enzyme by a cell.
No, but I could accept it for purposes of discussion if you also were willing to accept your original critique of Axe’s paper is equally flawed.
That’s the purpose of this thread. I’m asking if anybody has suggested an actual evolutionary trajectory from the actual, ancient enzyme to the modern beta-lactamase. Because if your claim is that beta-lactamase activity evolved after the fold did, than in that case the probability from the ab screen studies that were brought up might be actually relevant. Do you see what I mean? If you start with the fold but it’s catalyzing some other reaction and evolved to perform the beta-lactamase function, than those probabilities might be relevant and I could potentially even agree. We have lots of examples of promiscuous enzymes and novel enzymatic function. But that still doesn’t explain the origin of the fold.
If you claim that the function evolved before the fold, than I don’t see why the ab screen study numbers are relevant at all. In Darwinian terms, the modern enzyme would be expected to be the top of a fitness peak which started with beta-lactamase function but no fold. And it just doesn’t seem likely to me that the fitness landscape with 50+ orders of magnitude more local peaks for beta-lactamase function than protein folds would just happen to find the base of a peak that has a complex fold at the top. You would be just assuming that all enzyme functions exist at the base of fitness peaks which have complex, highly stable folds at the top. That’s just assuming the point under discussion.
Then discussion will not be possible as your position isn’t based on evidence or reason.
Could someone here provide a definition of “fold” sufficient to determine when two proteins have the same or different folds? How many folds are there in, say, SWISSPROT?
Discussion is also difficult when people try to pull one over on me, such as suggesting TEM-1 has peptidase activity when that has only been shown for P99.
I highly doubt the antibody will tell us anything about the origin of the beta-lactamase in bacteria since they had very different origins. What the antibody does tell us is that there are many possible targets for evolving that enzyme activity, contrary to the claims made by Axe.
Random peptide sequences will form folds, so I’m not sure why you find protein folds so improbable.
They have, yes. The first paper I linked in this thread has a figure depicting the phylogenetic tree where the origin of serine beta-lactamases derives from DD-peptidases.
Yes your question makes sense. But I’m not claiming that I know the history of when the activity evolved in relation to the specific fold. It is entirely possible that the class of enzymes that predate the evolution of specifically antibody-degrading enzymes, still had promiscuous catalytic capacities even before the substrates they would eventually evolve to act on, had themselves evolved.
The evidence implies that the class of enzymes that predate the serine beta-lactamase superfamily, the DD-peptidases, were back then and still are involved in cell wall (mureine) biosynthesis. Structurally they seem to have some similarities to each other, though they are clearly not identical.
But this then raises the question of what exactly defines a protein fold. How different must one protein’s structure be from another, before we start to think we should classify it as a different fold?
The evidence actually implies that a considerable level of fold-change has occurred over this evolutionary history, both within the serine beta-lactamase superfamily, and in it’s relation to other superfamilies such as the DD-peptidases.
Here’s a typical class A beta-lactamase:
From Medeiros AA. Evolution and dissemination of beta-lactamases accelerated by generations of beta-lactam antibiotics. Clin Infect Dis . 1997;24 Suppl 1:S19-S45.
Here’s a class C:
While I can certainly see some considerable similarities, I can also see that they are clearly not identical. Yet they are both classified as part of the serine beta-lactamase superfamily.
The DD-peptidases are even more different while still retaining some structural similarities.
Nobody claims that one enzyme evolving into another necessarily explains the origin of the fold, at least if they really do share the “same” fold to some degree.
But this actually highlights part of the problem here, which is that you have to first decide, for yourself, inside your own head, at what point it makes sense to say that one fold originates, while gradually transforming from another?
After all, if that really is how it originated, from another fold, gradually and incrementally, then the decision is in some sense arbitrary. There isn’t an objectively right answer about where exactly the border between two folds “should be”, even though it may be an objective fact that it did evolve gradually from one structure into another.
[quote=“BenKissling, post:31, topic:11233, full:true”]
Seriously? @BenKissling, you are putting words in Axe’s mouth, claims that a plain reading of his 2004 paper will show to be false, and you try to defend your assertions?
Also, no one - not Axe, not @Agauger, not @bjmiller, not any other ID proponent - has ever really addressed my criticisms of the ID take on Axe’s 2004 paper. Why I should accept the false claim that my critique is flawed escapes me. Maybe you can help me understand why you ask me to accept false assertions.
Doesn’t sound random to me:
The authors admit this is not an actual random sampling of the entire sequence space. Nor did they really test for tertiary structure. They only tested for secondary structure and “collapse”, which just means that it wasn’t strung out. That doesn’t mean it had a stable tertiary structure or anything resembling a wild type fold.