Optima in Evolution

[Mercer (John Mercer)]

@T_aquaticus

From what I can see of MYH7, this protein is shared with fungi, so you would need to go the common ancestor of fungi and animals, and then start tracking mutations through the entire phylogeny leading to humans.

Not an ortholog, but multiple “unconventional” myosins that are homologs of MYH7.

Here’s a great paper on the family:
https://www.nature.com/articles/nature03949

[Giltil (Gilbert Thill)]

@Giltil

IOW, for most proteins, being able to adopt folded structures is a necessary condition for performing a complexfunction, but it is not sufficient.

So do I understand you correctly here, to be saying that even if we could somehow assemble a substantial library of random proteins that nevertheless fold into some structure, we’d still be very unlikely to find a biologically useful function?

In other words, that not only are folding proteins in general rare among protein all sequences, but even among the minority of protein sequence that do fold, biologically useful functions are rare too?

That’s pretty much correct in the sense that the target space for folded proteins is probably bigger that the target space of natural functional proteins. But how much bigger, I don’t know.

[T_aquaticus]

@Mercer

Not an ortholog, but multiple “unconventional” myosins that are homologs of MYH7.

Protein domain paralogs might be a good description. Good point.

[Mercer (John Mercer)]

@Giltil

That’s pretty much correct in the sense that the target space for folded proteins is probably bigger that the target space of natural functional proteins. But how much bigger, I don’t know.

So you didn’t even read the paper you cited.

If you don’t know, why are you making such grandiose claims?

[Rumraket (Mikkel R.)]

@Giltil

@Rumraket

So do I understand you correctly here, to be saying that even if we could somehow assemble a substantial library of random proteins that nevertheless fold into some structure, we’d still be very unlikely to find a biologically useful function?

In other words, that not only are folding proteins in general rare among protein all sequences, but even among the minority of protein sequence that do fold, biologically useful functions are rare too?

That’s pretty much correct in the sense that the target space for folded proteins is probably bigger that the target space of natural functional proteins. But how much bigger, I don’t know.

Okay. But I guess you are not aware, then, that scientists have tested that in multiple concrete experiments? They have generated large libraries of variants of folding proteins, in total ignorance of whether they will have any useful functions, put them into biological organisms, and tested to see how they work.

Here’s a couple papers on such experiments where they were tested for in-vivo beneficial functions in living cells:
Digianantonio KM, Korolev M, Hecht MH. A Non-natural Protein Rescues Cells Deleted for a Key Enzyme in Central Metabolism. ACS Synth Biol. 2017 Apr 21;6(4):694-700. DOI:
10.1021/acssynbio.6b00336

Digianantonio KM, Hecht MH. A protein constructed de novo enables cell growth by altering gene regulation. Proc Natl Acad Sci U S A. 2016 Mar 1;113(9):2400-5. DOI:10.1073/pnas.1600566113

These papers are interesting because they highlight the unpredictable nature of evolution. Key genes involved in complex biological functions inside living cells are deleted, and large libraries of random but folding proteins are screened to see if any of these proteins are able to functionally take the place of the deleted genes.

You might naively predict that, in so far as they are able to, they do it by taking over and performing the function of the deleted genes. But it turns out instead that they function in entirely different ways. The proteins being tested turns out to function as trancription initiators (which means they either have to interact directly with DNA, and/or and bind other transcription factors for example by suppressing other inhibitors, and/or bind to RNA polymerase) that upregulate expression of certain metabolic enzymes with low-level promiscous side-reactions that they are normally not selected to perform. So this in turn proves two things:

1. First, it proves that already existing proteins are often times functionally promiscous, which means one protein can have many functions they have normally not evolved or been selected to perform, which they can nevertheless perform at some low level, and under the right conditions those functions can become adaptive, which would then provide the basis for further enhancement of those functions by selection.
2. Second, it proves that among folding proteins in general, it cannot be the case that biologically useful functions are too unlikely to evolve.

There are many other such experiments, where naively designed proteins, made only to be able to fold, have been tested for in vitro functions, such as small molecule binding and enzymatic activities. Here’s a couple of those:
Cherny I, Korolev M, Koehler AN, Hecht MH. Proteins from an unevolved library of de novo designed sequences bind a range of small molecules. ACS Synth Biol. DOI: 10.1021/sb200018e

Patel SC, Bradley LH, Jinadasa SP, Hecht MH. Cofactor binding and enzymatic activity in an unevolved superfamily of de novo designed 4-helix bundle proteins. Protein Sci. 2009 Jul;18(7):1388-400. DOI: 10.1002/pro.147

These papers of course demonstrates what was already implied above by the fact that native proteins in living organisms usually have many low-level side-functions(both enzymatic and otherwise) they haven’t evolved to perform, yet which nevertheless natively exist as a capacity of their sequence and structure. This also explains why so much of protein evolution has happened through gene-duplication, as many proteins with multiple native functions have been duplicated and repurposed to enhance those functions as they became beneficial under the right circumstances.

There are also papers on random protein sequences that have not been explicitly designed to fold, being tested for adaptive biological functions in real living organisms. Such as this one:
Knopp M, Gudmundsdottir JS, Nilsson T, König F, Warsi O, Rajer F, Ädelroth P, Andersson DI. De Novo Emergence of Peptides That Confer Antibiotic Resistance. mBio. 2019 Jun 4;10(3). pii: e00837-19. DOI: 10.1128/mBio.00837-19

This last one is interesting because they screen a library of a few hundred million small peptides, some of which are too small to yield actual protein folds(thus proving that you don’t need folds for biologically useful functions) and only really form secondary structures, such as single sheets or helices. They find multiple small proteins in the 22-25 amino acid range that function as membrane channels. These membrane channels have the function of depolarizing the bacterial membrane which reduces antibiotic uptake, leading to an almost 50-fold increase in the amount of antibiotic tolerated by the organism.

This of course also proves that a protein doesn’t need to fold to be functional, and that biologically relevant and useful functions can’t be too rare to evolve.

[Rumraket (Mikkel R.)]

@Giltil
So given that the evidence is against your belief, what’s next?

[Rumraket (Mikkel R.)]

@Giltil
You there?

[Faizal_Ali]

@colewd

This is very disappointing that you don’t realize this. Your understanding of the issues is not what I thought it was.

What a disgusting thing to say.